Journal of the American Oriental Society

JOURNAL

OF THE

AMERICAN ORIENTAL SOCIETY.

EDITED BY

E. WASHBUKN HOPKINS, AND CHARLES C. TORREY

Professor in Yale University,
New Haven.

Professor in Yale University,
New Haven.

TWENTY-SEVENTH VOLUME.

SECOND HALF.

THE AMERICAN ORIENTAL SOCIETY,

NEW HAVEN, CONNECTICUT, U. S. A.
MC M VII

A copy of this volume, posture 'paid, may be obtained any-
where within the limits of the Universal Postal Union, by
sending a Postal Order for two dollars and flfty cents, or its
equivalent, to The American Oriental Society, New Haven,
Connecticut, United States of America.

According- to the conversion-tables used in the United States
money-order system as the basis of international money-
orders, two dollars and flfty cents ($2.50) = 10 shillings and 3
pence = 10 marks and 30 pfennigs = 12 francs or lire and 70
centimes = 9 kroner and 25 ore = 6 florins and 9 cents Nether-
landish.

[This volume is for July-December, 1906. Issued March 12, 1907, in an edi-
tion of 500 copies.]

\

Copyright, 1907, by
THE AMERICAN ORIENTAL SOCIETY.

V

.27

The Tuttle, Moreliouse & Taylor Pros*.

CONTENTS

OF

TWENTY-SEVENTH VOLUME,
SECOND HALF.

Patre

A DISTINGUISHED FAMILY OF FATIMIDE CADIS (AL-NU'MAN) IN THE TENTH
CENTURY. — By RICHARD GOTTHEIL, Professor in Columbia Univer-
sity, New York City _____ .............. . ............. . .......... 217

SUPPLEMENT TO THE OLD-BABYLONIAN VOCABULARIES. — By R. J. LAU,

Ph.D., Columbia University, New York City ..................... 297

ABEL (^^JiJ) IN THE BIBLE. — By R. J. LAU, Ph.D., Columbia Univer-
t sity, New York City .......... . ........ .......... . ........ _____ 301

THE PI'LEL IN HEBREW. — By Louis B. WOLFENSON, Johns Hopkins

University, Baltimore, Md ...................................... 303

CONTRIBUTIONS TO COMPARATIVE PHILIPPINE GRAMMAR. — By FRANK R.

BLAKE, Ph.D., Johns Hopkins University, Baltimore, Md .......... 317

NOTES ON SOME PALMYBENE TESSERAE.— By HANS H. SPOER, Ph.D.,

Jerusalem, Syria. . ................ ________ .................... 397

THREE OBJECTS IN THE COLLECTION OF MR. HERBERT CLARK, OP JERU-
SALEM. — By GEORGE A. BARTON, Professor in Bryn Mawr College,
Bryn Mawr, Pa ........ _______________ .......................... 400

STUDIES OF SANSKRIT WORDS. — By EDWIN W. FAY, Professor in the Uni-

versity of Texas, Austin, Texas .................................. 402

NOTES ON THE MRCCHAKATIKA. — By ARTHUR W. RYDER, Ph.D., The Uni-

versity of California, Berkeley, Cal ........ - ...................... 418

THE BUDDHISTIC RULE AGAINST EATING MEAT. — By E. WASHBURN HOP-

KINS, Professor in Yale University, New Haven, Conn ............. -I •*»•"»

Proceedings at New Haven, Conn, April 17 and*18, 1906..

Attendance ....... . ..................................... - 465

Correspondence .............................

Necrology ............................... ........... - :

Report of the Treasurer. ....................... ................ 467

IV

Page
Report of the Librarian 468

Report of the Editors .469

Members elected - 469

Officers elected - 470

Report of the Directors 472

List of Members, 1906 474

List of exchanges 483

List and prices of publications 491

Notice to contributors and general notices 492

JOURNAL

AMERICAN ORIENTAL SOCIETY.

A Distinguished Family of Fatimide Cadis (al-Nu'mdn)
in the Tenth Century. — By RICHARD GOTTHEIL, Professor
in Columbia University, New York City.

I. INTRODUCTION.

IN the whole of Mohammedan history there are few epochs
quite as interesting as that during which the Shi 'a propaganda
manifested itself politically in' Egypt, maintaining 'there for
more than 200 years a kingdom which was a center of com-
mercial and literary activity. The religious side of this propa-
ganda was kept alive by the usual Alid tergiversations, and
from out of this upbuilding came much of the turmoil in which
Druse and Ismailian pretensions were hatched.

It seems to have been a somewhat simple matter for the
people of Egypt to pass from one system to another. They
were willing to take their religion as it was given to them, and
at no time do they seem to have thought with Goethe :

" Was du ererbt von deinen Vatern hast,
Erwirb es, um es zu besitzen."

In spite of the large Coptic element in the population,1 it had
not been too difficult a task to impress the faith of the prophet
upon the land of the Pharaohs. Egypt is the classic home of
the corvee ; and, whether used by an old Pharaoh in dragging
his statue to the place of its permanent situation, or by 'Amr ibn

1 This has been excellently set forth in Butler's Arabic Conquest of
Egypt, Oxford, 1902.
VOL. xxvji. 15

218 E. J. H. Gottheil, [1906.

al-'Asi in re-cutting the canal that once joined lower Egypt to
the Red Sea, or by IsmaiLPasha in helping the French to build
a Suez Canal, it shows a more. than ordinary apathy on the part
of the inhabitants, and a singular willingness to acquiesce quietly
in the stings and goads of fortune. In the same manner, it
does not seem to have been too difficult for them to pass from
the Sunnite faith to the Shi'ite (if faith it may be called), when
Jauhar al-Ka'id conquered the country in 969 for his master
ul-Mu'izz ; and they were as ready to fall back again upon the
Sunna Avhen the Kurd Saladin, in September, 1171, caused the
Khutbah to be pronounced in the name of the Abbasid caliph,
al-Mustadi.

One reason for the ease with which these changes were effected
must be found in the small difference it made to the people at
large whether in the official utterances Ali was blessed or cursed.
That was food for the theologians and a tid-bit for the jurists.
The lower classes had to live their every-day and humdrum
life as they had done in the past ; and the differences between
Sunnite and Shi'ite actual practice seems to have been small —
to us they appear infinitesimal.1 The geographical writer
al-Mukaddasi has an interesting passage on the observances
peculiar to the Fatimides.2 He says : " There are three classes of
Fatimide peculiarities. The first is one in wThich the (orthodox)
Imams were also divided, as the long inserted or supererogatory
prayer 3 at the morning devotions and the audible recitation of

1 This is due to the fact that the Shi'a system was developed at a time
when the other and canonical legal systems were already in existence.
Both the Sunna and the Shi'a, as regards their religious practices, are
built up from one and the same basis. See Von Kreiner, Culturge-
schichte, vol. i, p. 501 ; id. Herrschende Ideen, p. 389.

2 In de Goeje, Bibliotheca Geographorum Arabicorum, vol. iii, p.
237. 16. On the o-A^M J^JCJ ^JC juo see Ibn Khaldun, Mukdddamdt (ed.
Beirut, 1886), p. 390. A list of works on Shi'a Fikh is given by al-Nadlm,
Fihrist, p. 219. On some other and equally minor points of difference,
see the end of the poem by Da'ud ibn ' Umar al Baslr al-Antakl in his
o'tj-**^ ..wjoyj' ^ jLx^published by Goldziher, Beitrage zur Litera-

turgeschichte der Si'a in Sitz. Her. der Wiener Akad., vol. Ixxviii, p. 520,
and compare Tornauw, Le Droit Musulman, Paris, 1860, p. 24.
J De Sacy (Chrestomathie Arabe, vol. i, p. 162), says that the ci>»jvj is

the prayer containing the formula ,j,JCjljj dU Li! ; but see the tradition

Vol. xxvii.] A Distinguished Family of Fatimide Cadis. 219

the basmallfih,1 the witr* which goes with the rak'a, and the
like. Their second peculiarity is to return to some of the
observances of former generations, as the double repetition of the

cited in Lane, col. 2566 cu^xAj! ^So «i>^L<aJf jL*ii!, and al-Bukharl,
Sahlh, ed. Krehl, vol. i, p. 204, s.v. v^xftj! voL ; al-Shirazi, al-Tnnbih,
(ed. A. W. T. Juynboll, Leiden, 1879), p. 24. 21; al-Sha'rani, Kashf
al-Ghumma, Cairo, 1281, vol. i, p. 85. It is evident that the «i>*JLi
is a sort of supererogatory prayer (the Mohammadans call such
Jot«j) inserted between the rak' as ( e «-$yJ! ..yjo /*^^ tXaJ C>XAJ)

The silent prayer between the rak'as is called p-Le4> (Goldziher, Mvham-
medanische Studien, vol. ii, p. 252). A. Querry, Droit Musulman, vol.
i, p. 81, calls it "le recueillement," while Tornauw (1. c., p. 57) explains
it thus : " le Kenut, qui consiste a elever les bras apres 1'accomplissement
des practiques mukerrenot et k repeter des interjections fervantes. Le
Kenut n'est point obligatoire." See, also, Hughes, Dictionary of Islam,
pp. 101. 1 ; 482. 2. Curiously enough, Muhammad ibn Ahmad al-Khu-
warazml in his Mafatlh al-'Ulum (ed-. Van Vloten, 1895), p. 11, says

j>yM pita ^yJLH !

1 See the traditions on this point in al-Bukhari, vol. i, pp. 197, 198, 201,
and al-Nawawi, as cited by Goldziher, Beitrage, pp. 457, 522, and in Ibn
Sa'd, vol. v. (Leiden, 1905) p. 266 (when Mohammed recited the first
Sura he was not heard to add the basmallah. Asked about this, he

answered: L£J ^>*.^4- LgJ\«.*J «J). Until the year 253 A. H. the

basmallah was recited aloud in Fustat ; then a change was made ; but
al-Jauhar reintroduced the older practice in 362 A. H. ; see de Sacy, I. c.,
vol. i, p. 162. The Shafi'ite practice was in this respect, as in so many
others, in consonance with that of the Shi 'a. See the quotation from
Abu-1-Fidaon p. 220, n. 3; and Abu al Nakib al-Tunturl -c^L^I -x2JC^X«

(Kazan, 1899), p. At. Al-Zamakhshari (al-Kashshaf, ed. Lees, vol. i, p.
5), has an interesting note upon the different usage in this respect.
According to him, the difference depended upon the question whether
the basmallah was or was not an integral part of the Sura; the "readers "

(olli) of Medina, Basra and Damascus held that it was not, and there-
fore did not read it aloud when it occurred in a prayer ( ^si ^ dU jJj
5 JLoJ! ,-3 jv^cX^t Lg->) ; but those of Mecca and Kufa did. See,
also, al-Baidawi, vol. i, p. 3.

2 The j'« is "a prayer accompanied by an uneven number of rak'as —
from one up to eleven. See al-Shirazi, al-Tanbih, p. 27. 5.

220 R. J. H. Gottheil, [1906.

ikama which the Banu Umayya had reduced to one ;' the wear-
ing of white, which the Banu al-' Abbas had changed to black.8
The third peculiarity is to follow such customs as the Imams
had indeed not prohibited, though they had not been known

before this time : e. g., to iise the expression s^JLoJI J^c. ~rs>.
" Come to prayer ! " in the adhan ; 3 to celebrate that day as
the first of the month on whch the new moon is sighted ; 4 and
to accompany the prayer said at an eclipse with five rak'as and
two sujuds to every rak'a."5

The religious and juridic rite originally followed by the
Mohammedans in Egypt had been that of Malik ibn Anas ;

1 The &oUsl is the call to prayer which immediately follows the ^f jl .
See Dozy, vol. ft, p. 424 ; Tornauvv, Le Droit Musulman, p. 57 : " Apres
le azon on prononce 1'ekomeh. Ce sont les memes paroles: Allah Akber —
seulement dans 1'ekomeh on ne jette 1'exclamation que deux fois, tandis
qu'on la repete quatre fois dans 1'azon." A. Querry, Droit Musulman,
vol. i, p. 66.

2 See the material upon the various colors affected by different Muham- .
madan parties which I have collected in ZA., vol. xiii, p. 194, note
1, and vol. xiv, p. 223, note 7. The change to white dress was immedi-
ately introduced upon al-Jauhar's gaining possession of the mosque of
'Amr (lOjuCxJi *x>L4-0 iQ Fos^at- See Stanley Lane-Poole, History
of Egypt, p. 103 and the following note.

3 According to Abu-1-Fida the expression was

r
He says (vol. ii, p. 498) ^xx^+ia.. «**j jiLLu/ ,.j.x> ^*$\ ^jL*.2» _9«

t -j

dLJ jo ^tyJtJ! (v°\^\ ^ s JOLJ (jc>! jvJ . J^»jJI .*=*>

' The same account

is found in Ibn Sa'd, ed. Tallqvist, p. 77. See, also, JA. 1836, 3, p. 57,
Ibn Khallikan, vol. i, p. 344, and Ibn Khaldun, Kitab al'Ibar, vol. iv,
p. t*A : and especially al-Makrizi, Khitat, vol. ii, pp. 340, 342, and the
full account, ib. p. 269.

4 See de Sacy, Chrestomathie Arabe, vol. i, p. 161.

5 On the Sunnite practice, see al-Bukhari, 1. c., vol. i, pp. PvP and HP

and Muhammad ibn 'All Siddlk Hasan c «-Aj — . ^XiJ *^\Jt_H ^-AJ

*'~»JI , vol. i, p. 221. On the different customs, see al-Sha'ram, Kitab
al-Mlzdn, vol. i. p. tvP.

Vol. xxvii.] A Distinguished Family of Fatimide Cadis. 221

but when in 813 Muhammad al-Shafi'I came to settle definitely
in Fostat, his legal doctrines commenced to gain prevalence1 and
they remained prevalent until in the 16th century the Turks
introduced the Hanafite system.2 Upon the arrival of al-.M iri/./.,
it was natural that this should be changed. But the Fatirnides
seem to have been rather large-minded in this respect, if not in
others.3 They were sufficiently latitudinarian to allow all the
four forms of Mohammedan canon law to be taught in the
schools and to be used by the adherents of different parties.4
From the time of al-Mustansir on, we read of Shafi'ite doctors
being appointed to the chief cadiship.5 Al-Mu'izz himself seems
to have gone slowly in forcing upon the country Shi'ite prac-
tices. The cadi whom he found in office, Abu Tahir, received
permission to continue his functions. Probably he fitted him-
self conveniently into the changed circumstances ; for it is
related that he came to Alexandria and greeted the new caliph
in a somewhat ostentatious manner. Al-Makrizi is quite explicit
in his statement that Shi'ite law was first taught in Cairo by the
son of the Fatimide cadi whom al-Mu'izz had brought with him.
"Fatimide law," he says, "according to Shi'ite doctrine was
first taught at the Azhar in Safar 365 (975), when' 'AH Ibn
al-ISTu'man, the cadi, sat in the Cairo mosque known as the
Azhar and dictated a compendium of law composed by his
father for the Shi'ites."6 This work was called al-Iktisar.7

1 al-Makrizi, Khitat, vol. ii, p. 334.

2 See Snouck-Hurgronje in ZDMG., vol. liii, p. 134.

3 See Guyard in JA., 1877, 1, p. 335.

4 In the year 425 A.H. four cadis were appointed: An Imami, an
Isma'ili, a Malikite and a Shafi'ite. See al-Makrizi, Khitat, vol. ii, p. 343
et seq.

6 The Shi'a naturally looked with more favour upon the Shafi'ite
system, because of the position the latter took in regard to the use of
the (j^Lk.* . See Goldziher, Beitrdge, pp. 485, 500.

> a favorite designation with which the Shi'a glorified

its political leaders. They, therefore, speak of the
see Goldziher in ZDMG., vol. xxxvi, p. 279.

1 I have the quotation from Mustafa Bairam's xJL«x on the Azhar

Mosque (Cairo, 1321 A. H.), p. 23. [See al-KhiM, vol. ii, p. 841.] The
title of this work is said by Ibn Khallikan (vol. iii, p. 565) to have been
' al-Intisdr;" see further on p. 228.

222 J?. J. J?. Gottheil, [1906.

Ya'kiib ibn Killis (the renegade Jew and the first Fatimide
vizier1) went further than did his master. During the reign of
al-'Aziz, the son of al-Mu'i«z, the Caliph compelled — as al-Mak-
rlzl also tells us — all the poets, philosophers, and learned men
to come to his house and listen to t>he exposition of a little book
that Ibn Killis had put together containing whatever of Ismaili
canon law he had heard from al-Mu'izz andal-'Aziz. On Tues-
days and on Fridays he was accustomed to hold a special levee,
at which he expounded Fatimide theories. The caliph made
attendance at these levees compulsory upon the learned men and
the doctors ; a special building being erected for that purpose
next to the Azhar. During the reign of al-Hakim, the people
were in such dread of their ruler that they joined the Shi 'a in
large numbers and 'Abd al-'Aziz al-Nu'man had to hold daily
sessions, at which the initiated were received.'2 The theologians
had evidently gained the upper hand ; and how stringent the
spirit was liable to become may be seen from the fact that in
the year 381 (991) a man was actually driven from the city
because a copy of Malik's al-Muwatta had been found in his
possession.3 In the year 380 (990) a special Jami' — called
al-Jami' al-Hakimi was erected for the benefit of the Shi'a pro-
paganda ; but it was not finished until the year 403 (1012). 4

As all Mohammedan law is really canonical law, the com-
mander of the ship of state had to depend very largely upon
his steersman at the helm. That steersman was usually and
naturally the cadi, and the cadi at first had a position second
only to that of the caliph himself. If al-Makrizi is to be cred-
ited, al-Mu'izz had no vizier at all ; and the duty of spreading
Shi 'a doctrine and of consolidating Shi 'a practices devolved
upon the cadi. The position that he held was therefore an impor-
tant one ; and, in addition, at times the superintendency of the
mint and of the bureau of weights and measures was also in
his hands. After a while the cadi also became the chief

1 al-Makrizi in Jamaleddinni Togri-Bardii Annales, ed. J. E. Carlyle
(Cantab. 1792), Notes, p. 5.

2 De Sacy, Les Druses, p. ccxi.

'Mustafa Bairam, 1. c., p. 23, al-Khitat, vol. ii, p. 341.
4 al-Makrizi, Khtfat, vol. ii, p. 277 ; Van Berchem, Corpus Inscrip-
tionum Arabicarum, p. 50.

Vol. xxvii.] A Distinguished Family of Fatimide Cadis. 223

preacher.1 The holders of the office must therefore have been
men of some significance, and their history is closely intertwined
with that of the country itself. August Milller, in speaking of
the Barmecides, and the services that they rendered to the
Abbfisid caliphate of Bagdad, calls attention to the fact that

1 Upon the various functions attributed to the cadi in addition to the
judgeship, see the instructive remarks of Ibn Khaldun, Mukad$amat,
p. HI** below. Ahmad ibn 'All al-Kalkashandi, in his work on the
geography and administration of Egypt, (at least in the part translated
by Wiistenfeld in Abh. der Konigl. Gesell. der Wiss. Gottingen, 1879, p.
184) speaks only of the surveillance of the markets being at times in his
hands. Happily, the whole of this informing work is in course of pub-
lication by the Khedivial Library in Cairo. The importance of the
"Chief Preacher" in the Fatimide period is justly brought out by
al-Makrizi (al-Khitat, vol. i, p. 390 ; see, also, De Sacy, 1. c., vol. i, p. 140):
upon him rested a good part of the onus to propagate Shi'a doctrines.
Al-Kalkashandi seems to know nothing of the union of the offices of

.<o\J> and sLfcjJ! _,c!i> : but al-Makrizi has the following :

o

»J JLftxs Lvcta ^IsYI u>oju U

~

sLejJf .c!t>« • Theoretically, it was the vizier to whom the func-
tions of the cadi belonged (see Mawardi, Const itutiones politicae, ed. R.
Enger, Bonn 1853, p. 39, 1); if he was unwilling to exercise the functions
he could appoint deputies. This must be the meaning of al-Makrizi

(Khitat, vol. i, p. 403) : ^ 13! *j! »J jJ!

ibLo !^Ls*s LoaJ! jJUb «jU ^-ax**/ . But historically,

the union of the two offices (viziership and cadiship) occurred only in a
few cases; al-Kalkashandi, in his account of the wazlr (1. c., p. 181)
knows nothing of it. Ahmad ibn Sa'd al-DIn al-Ghumri in his Sj.x~k j
,*^Lc^! (Paris Ms. 1850) mentions the cases of Ahmad ibn Zakariya and

Ibrahim ibn Kudaina. Muhammad ibn Ahmad ibn lyas recalls that
al-Yazuri, at the time of the Fatimid al-Mustansir, filled both offices :

According to al-Shirazi, the ultimate authority in the appointing of
a cadi was vested in the spiritual or virtual head of the community :

p. 313, 3).

224 R. J. H. Gottheil, [1906.

for more than fifty years this family was in the service of
the state. He adds: "Das ist meines Wissens sonst tiber-
haupt nicht und anderswo selten genug dagewesen."1 It is
therefore not without interest to see that in the early years of
the Fatimides, and for a term covering more than eighty years,
the office of cadi was held (with periods of intermission) by
members of one and the same family, named al-Nu'man ; and
I have tried in the following paper to reconstruct the history of
this family from both printed and manuscript sources.

In addition to the individual biographies of cadis in such
dictionaries as that of Ibn Khallikan and its continuation by
Muhammad Ibn Shakir al-Kutubi,2 the history of the cadis in
the chief Islamic centers formed a special branch of Mohamme-
dan biographical science. In his chapter on 'Ilm al-Ta'rlkh,3
Hajl Khalifa divides this science into the following categories:
1, the general history of the cadis ; 2, the history of the cadis
of Egypt ; 3, the history of the cadis of Bagdad; 4, the history
of the cadis of Basra; 5, the history of the cadis of Cordova;
and 6, the history of the cadis of Damascus.

The history of the Egyptian cadis seems first to have been
written by Abu 'Umar Muhammad ibn Yusuf ibn Ya'kub al-
Kindi down to the year 246 A. H. (860). " This was continued by
Abu Muhammad Hasan ibn Ibrahim ibn Zulak, who carried it
down to and through the biography of Muhammad ibn al-Nu'man
(386 A. H., i. e. 996). Hajl Khalifa says that an appendix to
this work of Ibn Ziilak was written by Shihab al-Dln Ahmad ibn
'All ibn Hajar5 up to the year 852 A. H. (1448) under the title

1 Der Islam, vol. 1, p. 465.

2 Fawdt al-Wafaydt, Bulak, 1283 and 1289 A. H.

3 Ed. Fliigel, vol. ii, p. 97.

4 A Ms. of this work is in the British Museum ; see de Goeje in ZDMG.
vol. 1. p. 741. Al-Kindfs LgJlSL^ii. _^ix> *_}Lx5^was published in 1896
by J. Ostrupp, (Bulletin de VAcademie Royale, Copenhagen, 1896, No. 4),
who has made it probable that this little tract of Abu 'Urnar was pub-
lished by his son 'Umar al-Kindi.

5 Abu-1-Fadi Ahmad ibn 'All ibn Muhammad ibn Hajar was born in
Ascalon (al-'Askalam) in 1372 and officiated as Chief Cadi in Cairo from
1424 to 1449. This will explain his interest in the history of his prede-
cessors in office. He was a most fruitful writer on hadith, and biog-
raphy — as well as something of a poet. See a list of his works in

Vol. xxvii.] A Distinguished Family of Fatimide Cadis. 225

^e »Uii ^ jsa$\ £jj . This work of Ibn Hajar seems to

be more than a mere appendix. It is really a biographical dic-
tionary, arranged in alphabetical order, and probably contains
all the data to be found in the preceding works.1 The .AV//4
al-Isr was continued by the author's pupil, Shams al-Dln
Muhammad ibn 'Abd al-Rahman al-Sakhawi, who entitled his
work »'j J'j ^U-UJI iU*j (That which is desired in regard to
the, etc.).2 It seems a pity that the work of Ibn Zuliik has not
come down to us ; al-Siyuti and Ibn Khallikiin evidently made
use of it, as they cite it several times.* Nor has Ibn Zulak's
other work, LgJJsLajj -o^o wsJ^', shared a better fate. I can

not believe that the Paris Ms. 18174 is really the work of so

Brockelmann, Gesch. der Arab. Lit., vol. ii, p. 67. A very full account
of the literary activity of Ibn Hajar can be found in his biography
written by Shams al-Din Muhammad ibn 'All al-Shahawi (Ms. Paris, No.
2105, fol. 191 b. et seq. — a voluminous work). Cf. also Quatremere,
Hist, des Sultans Mamlouks, vol. i, 2, pp. 209 et seq.

1 Those portions which deal with the family al-Nu'man will be found
below, both in text and in translation based upon the Paris Ms. No. 2149.
A second (and more correct) Ms. has lately been added to the same col-
lection from the library of the late Ch. Schefer. It is numbered 5893.
No. 2152 of the same collection, containing (jfl^i^sJCj 5.je!-,Jf p.-<vH
« -dx> sLxia )^=».t °y I"311 Hajar's grandson, Jamal al-Din

Yusuf ibn Shahm, is practically identical with the work of Ibn Hajar;
despite the author's remarks in the preface that his grandfather's work
was incomplete because death prevented him from making a thorough
revision. The two Mss. of Ibn Shahm that I have examined (Paris 2152
and Berlin 9819) are very similar and are evidently of the same prove-
nance. They are very correct, the Paris Ms. having been revised by the
author; though they are difficult of use for text-critical purposes, as the
diacritical points are wanting for the most 'part and the script is minute.

8 A mukhtasar of this was composed by 'All ibn al-Lu^f al-Shafi'i. See
Haji Khalifa, vol. iii, p. 473 ; vol. iv, p. 561.

3 See, also, Carl H. Becker, Beitrdgezur Geschichte Agyptens, i. p. 14.

) -H^ LgJl2L*ii« -»a>o ^j»Li' . On Ibn Zulak see Ibn Khal-

likan (de Slane's translation) vol. i, p. 388 — who mentions only his topo-
graphical description of Egypt and his history of the Egyptian cadis.
The latter, it is known, was merely a continuation of a work with the
same title by al-Kindi.

226 Jt. J. H. Gottheil, [1906.

important an authority as Ibn Zulak seems to have been. It is
hardly of more worth than Ms. 1816 of the same library, and
of which the compilers of the catalogue very properly say ' ' cet
opuscle ne'renferme que des fables."1 I may also mention Ms.
1819, which the catalogue describes as identical with Ms. 1817. s
At a later date Ahmad ibn Sa'd al-Din al-Ghumri al-Shafi'I
wrote a history of Egypt down to the year 1640 in double rajaz
verses, with the dates given in the form of chronograms.3 To
this he appended a list of the cadis. To write such and many
other things in verse was often an affectation of an Arab
writer. Even before the time of al-Ghumrl, a similar thing had
been done by one Abu 'Abdallah Muhammad ibn Daniyal
al-Mausili al-Khuza'I (died November, 1310), an oculist in Cairo,
dealing specifically with the cadis of Egypt in ninety-nine
verses of a like kind4; to which al-Siyuti added those who
had officiated from the time of Badr al-Jama'a up to his own
day (1481). 5 Probably more important than these works must
have been a history of Egypt written by Muhammad ibn Abi-

1 Catalogue, p. 330. LgJot. LjO.Uj^! ^uo JoLoJ

Another Ms. of this work is described in Fliigel, Die arabi-

ischen . . . Handschriften der k. k. Hofbibliothek zu Wien, vol. ii, p.
148. See, also, Blochet in Revue de VOrient Latin, vol. vi, p. 460. A
renewed examination of Paris Mss. 1816, 1817, 1818, 1819 and 1820 has
convinced me that not one of them can really claim to be the work of
Ibn Zulak. Nos. 1817, 1818, 1819 and 1820 exhibit practically one and
the same text; 1817 and 1819 going back to one and the same Ms. As
these discuss events as far down as the Ottoman invasion, it is impossi-
ble that 1817 is by Ibn Zulak. Nos. 1816 and 1818 are for a great part
merely a shorter and a longer recension of one and the same treatise.
No. 1816, fol. 45b mentions the 'Ukud al-Duriyyaofal-Jazzar, who died
in 1281 A.D. ! I hope to return to Ibn Zulak upon another occasion.

3 Brockelmann, 1. c., vol. ii, p. 297. There are some 9,000 verses in the
Berlin Ms. of this work. See Ahlwardt's Catalogue, No. 9831. I have
given, further on, that portion which deals with the al-Nu'man family,
taken from Paris Ms. No. 1850.

4 Brockelmann, 1. c., vol. ii, p. 8. They form the basis for Ibn Hajar's
Raf al-Isr, and are there cited in full.

6 Both are printed in al-Siyu^i's Hum al-Muhddara (Cairo, 1321), vol.
ii, pp. 117-121.

Vol. xxvii.] A Distinguished Family of Fatimide Cadis. 227

1-Kasim 'Ubaid Allah ibn Ahmad al-Musabbihl (976-1029).'
This work, entitled LgJ-jLoj^ ^ua ^IxiJ ^JuS 'is said to have
comprised some 26,000 pages, and is frequently quoted as one
of the best authorities ; but only an occasional part has remained
to testify to its worth. There are, of course, plentiful notices
about the Egyptian cadis in the monumental work of al-Mak-
rlzl (1364-1442; al-Khitat)', and al-Siyiiti in his Husn al-Muha-
dara has a special chapter, headed -«ax> »LdJ> /-^-2 The
material for this present study has been gotten chiefly from the
dictionary of the Egyptian cadis by Ibn Hajar, from the bio-
graphical dictionary of Ibn Khallikan, from the Khitat of al-
Makrizi, and from the short notices on the family to be found
in Ibn Khaldun's Kitab al-'Ibar, vol.. iv (p. 55). Ibn Khallikan
and Ibn Hajar have evidently used very much the same sources
— Ibn Zulak and al-Musabbihl.

II. THE FAMILY OF AL-NU'MAN.

Kaffir the Irshid had placed Abu Tahir Muhammad ibn
Ahmad ibn 'Abdallah al-Baghdadi al-Dihli al-Maliki in oflice
as cadi dm-ing the year 348 A.H. He remained cadi, -some say
for sixteen, others for eighteen years. When al-Mu'izz came to
Egypt on Sha'ban 23, 362 (May 29, 973), he brought with him
his own cadi, al-Nu'man.3 The father of al-Nu'man, Abu

1 His biography is given in Tallqvist, Fragmente des Ibn Sa'd, pp.
96-99 : 102-104 ; Ibn Khallikan (transl.) vol. iii, p. 87. He lived 976-1029.
See, also, Becker, I. c., i, pp. 16, 32 et al. It is well known that the
name is often found in Mss. as s^^+Jt. For the correct pronunci-

ation, in addition to the authorities quoted by Becker, I. c., p. 16, note 3,
see Ibn Khallikan, vol. iii, p. 90, and al-Dhahabi, al-Mushtabih, ed.
P. de Jong., 1881 p. fAt" .

3 Ed. Cairo, 1321 A.H., vol. ii, p. 95.

3 According to Abu-1-Mahasin (vol. ii, p. 488) al-Nu'man was origin-

ally a follower of the Hanifite School. aLftxxi! .

He is here designated as belonging to the Batiniyya ;
which,! believe, is merely the equivalent of the ordinary term

228 R. J. H. Gottheil, [1906.

Hanifa, was himself a well-known litterateur, who had died at
the advanced age of 104. Al-Nu'man had acted as cadi to the
army of the Fatimides on its journey from the land of the Ber-
bers ; but Abu Tahir came to Alexandria to meet al-Mu'izz, and
seems to have pleased al-Mu'izz so much that he was allowed
to continue in office.1 Ibn Nu'man had thus little to do ;
but occasionally he was called upon to revise the judgments
given by Abu Tahir. One of these cases must have occurred in
the year 974, for he died before the case was concluded, either
on Rajab 1, or on the last day of Jumada 363 A.H.2 Al-Nu'man
seems to have been a learned jurist. At first an adherent of
the school of Malik, at a later time he adapted himself to the
Shi'a teachings of his master, al-Mu'izz. To his first period
belongs a work v^sMtXtJI Jp&l o^.^~».t ^Lx5^ dealing with the
different principles upon which the various schools founded their
systems. He then became an ardent Fatimide, and placed his
pen in its service, writing a work ^lg.ft.oJI o^V-X^-t upon the dif-
ferences between the jurists, in which he defended the Shi'a

claims. To the same class belongs his Sj-c<XM ottXi'jf sr>Uc5'

an account of the first preaching of Fatimide doc-
trines. Two further juridic works written by him are mentioned :
jLub>.^H volxS' an(^ \L*2JC3^)t i^ jLx^" (The Triumph or Vindicator) ;
but to what subject they refer, we do not know ; and a poetic
piece i«>..^X.A.4~M (Choice Selection). Only one of his writings

Al-Shahrastam, al-Milalw-al-Nihal, ed. Cureton. p.

..vsi ,j«J«jLj *.#« .... auyLbUJt ; Muhammad ibn Ahmad
al-Khuwarazmi ..JLjiJI ^sxJ'Lax) (ed. G. Van Vloten) 1895, p. H :

1 Ibn Khallikan, vol. iii, p. 379 : Hum, vol. ii, p. 101.

2 See the case cited by Ibn 'Hajar. According to the latter, this
occurred on the 25th day of the month : but the name of the month is
not given.

3 Or, perhaps more correctly; 5«J&liiiJ! S..&JJ! ^-UiLXjt ^LxS" as

given by al-Makrizi in his Lflji+Jl ^UC^ . A small extract from this
work is given by Quatremere in JA. 1836, ii, p. 123. See, also, Brockel-
mann, I. c. vol. 1, p. 188.

Vol. xxvii]. A Distinguished Family of Fatimicle Cadis. 229

on law seems to have been spared ^-yJI JoLaJ _i x 1^.^11 _ ^xo
on the excellency of the prophet and the claims of All, which
may be the ^LAJ=»^! ^\..*S mentioned above ;' while of his
three polemical work against Abu Hanlfa, Malik, al-Shafi'I and
al-Suraij,2 one may be the ^LaXJ^t ^1x5^. He is also said to
have written a work on the " meritorious and disgraceful acts
(committed by the Arabian tribes)." He is praised by Ibn
Ziilak for his knowledge of the Koran, of Arabic poetry, philol-
ogy, pre-Islamic history, and jurisprudence. He left several
sons, two of whom followed him in the office of cadi.

Abu Tahir was evidently growing old, and al-Mu'izz gave
him as assistant or associate the son of al-Nu'man, Abu al-Hasan
'All. 'All was born in Rajab 328 (940), 3 probably in Mahdiyya,
the city in which the dynasty took its rise. He officiated in the
Jami' al-'Atik in old Cairo, while Abu Tahir had his own Majlis.
When al-Mu'izz died, in 365 A. H. (975), his successor al-'AzIz
confirmed the arrangement made by his father. In addition,
Abu al-Husain was appointed over the mint,4 an office often com-
mitted to the care of the cadi in those days, as well as over the
two mosques, probably the one in Fostat and the other in Cairo.
This double authority could naturally not last long, A case is
mentioned where Abu Tahir imprisoned certain persons, who
however appealed to 'All and were set free. As age and disease
crept over him, Abu Tahir had to be carried about ; and one
day he met the caliph at the Bab al-Diyafa5 and asked him to

1 Berlin Ms. No. 9662, though this seems to contain only an extract
from the larger work on " die Vorzilglichkeit des Profeten und die
Berechtigung Ali's auf die Nachf olgerschaf t. "

8 Fihrist, vol. 1, p. 213.

3 Ibn Khallikan, however, says Rabi' i. 329.

4 On the situation of the ^->««a.J! J& see al-Makrizi, al-Khijaf, vol. i,

pp. 406, 445 ; Ravisse, Essai zur Vhistoire et sur la topographic du Caire,
p. 76 ; P. Casanova, Histoire et Description de la Citadelle du Ca,ire,
p. 720.

5 The Bab al-Diyafa must have been near to the citadel. There was
a palace called Ddr al-Diyafa. See Casanova, La citadelle du Caire in
the Memoires of the Mission archeologique fran$aise du Caire, vol. vi,
p. 738, and Ibn lyas, _o.x» x5~>;Lj> ^lATiBulak 1311, A.H.), vol. i, pp.

147, 4; 310, 18. Ibn Khallikan says "near the Dar al-Sana'a" (al-
Makrizi, al-Khijaj, vol. ii, p. 178).

230 7?. J. H. Gottheil, [1906.

appoint liis son Abu al-'Ala as his substitute. This favor was
denied and after three days1 he was deposed and the sole cadi-
ship was given to Ali. It seems, however, that Ali's power was
at that time not quite complete. The friends and patrons of
Abu Tahir interceded for him, and he continued his functions in
his own house, perhaps revising the judgments given in other
courts. Ali was now publicly invested in the two mosques ; in
that of old Cairo it was his own brother Mohammed who read
out his diploma. His appointment was complete not only over
all Egypt, but over those countries also subject to the Egyptian
Caliph. In addition he was appointed chief preacher, inspector
of gold and silver, and controller of weights and measures.
Ali, himself, soon needed assistance. His brother Mohammed
was appointed to be his deputy in Tinnis, Damietta and Farama ;
while a second deputy was added in the person of al-Hasan ibn
Halil. The latter was a follower of al-Shafi'I ; but a special
order was given him to decide according to the Shi'a rite.2_

In the year 367 3 the caliph al-'Aziz was called to Syria to
quell the revolt of the Carmathians, whom his lieutenant, Jau-
har, had been unable to hold in check. He took his cadi, Ali,
with him, and Ali's brother Mohammed was appointed his sub-
stitute during his absence. Some enemies spread the report
that he had in reality been superseded ; and from where he
was with the army he was forced to send word to the prefect of
police, Hasan ibn al-Kasim, asking him to deny the rumor and
to strengthen the hand of his brother. Ali was upon very
friendly terms not only with al-'Aziz, but also with his vizier,
Jacob ibn Killis. It is said that he was the first to bear the
title "Chief Cadi" in his diploma.4 Before his time the title
had been granted only to the cadi at Bagdad. Ali died on

1 Ibn Khallikan has " two days."

2 See supra, p. 218, note 2.

3 Not 368, as Ibn Hajar has. See Wiistenfeld, Fatimiden Califen, p.
138-9 ; de Goeje, Memoire zur Us Carmathes du Bahrain (Leiden, 1886),
p. 192.

4 Husn, vol. ii, p. 101. He seems to have been recognized as chief
cadi, though the title was not given to him in his diploma. According
to Ibn Khallikan (vol. iv, p. 273), the celebrated Abu Yusuf Ya'kub
al-Ansarl, the author of the Kitdb al-Kharaj, was the first to have the
title

Vol. xxvii.] A Distinguished Family qf Fatimide Cadis. 231

Rajab the 6, 374 (Dec. 3, 984). Al-'Aziz, who was in camp at
al-Jubb, a plain near Cairo, where all sorts of assemblies
popular and military were held,1 came to the city, and himself
said the prayers over the corpse, which was then buried in the
Hamra." He was a well educated man, learned in jurispru-
dence, philology, polite literature and poetry.3 In fact, he was
something of a poet himself, and a few of his verses are cited
by al-Tha 'alibi in his Yatlmat al-Dahr, by Abu al-IIasan
al-Bakharzi,4 and by Ibn Zuliik.

Ali's brother, Abu 'Abdallah Muhammad, who had acted as
his substitute while he was in Syria, was formally appointed
cadi on Friday, Rajab 22, 374 — the office having been vacant
for 17 days on account of the sickness of the new cadi. Born
in the Maghrib, various stories are told that as a boy he had
been singled out by Al-Mu'izz for the position that he now occu-
pied. He must have been a man of some abilities to have held
so important a post. He was constitutionally weak-bodied and
was compelled to ride about in a palanquin. In this manner he
was carried to the camp of al-'Aziz for investiture ; and he was
even unable to be present when his son 'Abd al-'Aziz read out
his diploma in the Jami' al-'Atik at Fostat. In this diploma
he was appointed chief cadi over the whole of Egypt and the
Syrian possessions of the Fatimides ; he was also leader in
prayer, inspector of gold and silver, and controller of weights
and measures as his brother had been. Not being able to attend
to all his duties, he devolved some of them upon his nephew,
Abu 'Abdallah al-Husain ibn 'All, who was to hear cases in the
Jami' al-Hukiml. At the suggestion of the caliph himself, he
appointed his son, Abu al-Kasim 'Abd al-'Aziz to be his repre-
sentative in Alexandria. He stood high in favor with the govern-
ment ; his son, 'Abd al-'Aziz being married to the daughter of

M

1 Really Jubb 'Amira, Yakut, vol. ii, p. 18, 4 : ^jje \«>-5>J> »»£*-£ V^5

J'L-JLJL " U>! xJI Xj-o 5v#UJt . A ^rlt atfyj is mentioned by
\/ * ^_ '//'"/ /

al-Makrizi. See de Sacy, /. c., vol. i, p. 187.

2 Yakut, vol. ii, p. 333, 3 says simply ^ox>

3 It is remarkable that Ibn Hajar has nothing to say about his literary
attainments.

4 Brockelrnann, 1. c., vol. i, p. 252.

232 7?. J. If. Gottheil, [1906.

Jauhar the Kii'id, at a levee held by the caliph himself (Friday,
Jumfida 1, 375 = September 19, 985). It is even related that
upon one occasion (Jan. 5, 996) he accompanied the caliph into
the pulpit ; and when al-'Aziz died in 386 A. H. (996) Muham-
mad had the honor of washing his corpse.1 This natural Iv
excited the jealousy of the vizier Ibn Killis, who was afraid of
the growing power of the al-Nu'man family. Muhammad, how-
ever, succeeded in maintaining his position even under al-Hakim,
the successor of al-'Aziz ; being high in the favor of the eunuch
Barjawan, the tutor of the young prince. He was also quite
intent. to advance his own immediate family. In Jumada 1, 377
(Sept. 7, 987) he removed his nephew Abu 'Abdallah, to whom
he had given over some of his functions, and placed in his stead
his own son, 'Abd al-'Aziz, giving him authority to act as judge
on Mondays and Thursdays. In Muharram 383 (Feb. 993), he
increased the authority of 'Abd al-'Aziz by allowing his son
to sit as judge on every day.

This power of the cadi was distasteful also to the theologians
and the jurists. ; for he forced people to address him with the

title Ljt\A*u , "Our lord."! Ibn Hajar seems to refer to some
such dispute that occurred in the year 382. Muhammad had
appointed a certain Ja'far to publish the religious enactments in
the Jami' according to the Meccan rite.3 To this the Fakihs
objected ; but their objection was overruled in a summary
manner.

On account of his physical weakness he was unable to attend
to much of his work during his latter years. He is said to have
been of fine appearance, noble in his bearing and a ready giver
of alms. Ibn Zulak pays him the compliment of saying that
he knew of no cadi, either in Egypt or in 'Irak, who could be
called his equal. . Though learned in all the Moslem sciences,4

1 When the Caliph al-'Aziz felt his end approaching, he recommended
his son al-Hakim to Muhammad ibn al-Nu'man and to Abu Muhammad
al-Hasan ibn Ammar, the Amir al-Daula. See Ibn Khallikan, vol. iii,
p. 528.

2 On the use of this title, see Van Berchem, I. c., pp. 385, 386.

3 1 am not certain to what rite reference is made here — probably to that
of the Zaidite sect. The Sharifs of Mecca afterwards went over to
Shafi'i practices. " See Snouck-Hurgronje, Mekka, vol. ii, p. 251 f.

4 He lectured upon Shi'a law and doctrine. So many people crowded
to hear him upon one occasion that a number were killed in the crush.

Vol. xxvii.] A Distinguished Family of Fatimide Cadis. 233

he left no work behind him ; but he followed the Arabic custom
of writing poetry, a verse or two of which have been handed
down. Al-Musabbihi, however, did not think much of his
poetic talents. He died on Tuesday evening^ Safar the fourth,
389 (Jan. 25, 999), having been in office 14 years, 6 months and
10 days. Al-Hakim, himself, said the customary prayers over
his body. He was buried at first in his own house and then on
Ramadan the ninth (August) his corpse was transferred to the
Karafa cemetery.1 His palace was given to one of the friends
of al-Hakim, and all his possessions were sold in order to pay
the money of orphans and minors that had been deposited with
him.2

For some reason no chief cadi was immediately appointed to
succeed Mohammed. According to Ibn Hajar, who follows
al-Musabbihi,3 the interregnum lasted for seventeen days ;
according to Ibn Khalltkan for more than a month. On Safar
the 23,4 Abu 'Abdallah al-Husain, the son of 'All, who had held

Al-Makiizi, upon the authority of al-Musabbihi : aU^-JI^t .05

ge. oou j f^JU g JiS

-Xi.fi. c^ JU^ ^ cy+i L_>oLj auo y**1*? Xxa*- &J

xXJL jr?V«J^ r*-g-*-a^9 ^r*; (°r ^.fjjAftj) (al-Khitaf, vol. i, p. 390,
top, and de Sacy, 1. c., vol. i, p. 139).

1 To the south of Cairo. See Yakut, vol. iv, p. 48 ; al-Makrizi. Khi{a{,
vol. ii, p. 443: Rieu, Supplement to the Catalogue of Arabic Mss., p. 448;
Van Berchem, 1. c., p. 26.

2 It was customary to deposit in the chancellerie of the cadi moneys
belonging to orphans or to persons who were absent. See Mawardi, I. c.,
p. HA , in the chapter headed <>LoJLM RJ^« ^3 . Such moneys could
not be loaned out : though this was, of course, occasionally done under
pressure. Al-Makrizi relates one such incident in the life of Saladin.
In the year 590 A.H. he needed money for one of his campaigns; so
he forced the cadi Zain al-DIn to take 14,000 dinars that were in his
keeping and give them to him. See Blochet in Revue de I' orient Latin,
vol. ix, p. 76.

3 He says expressly that al-Husain was nominated on Safar 27.

4 And not on Rabi' 6, as Ibn Khallikan says — who seems to know very
little about this cadi.

VOL. xxvii. 16

234 7?. J. H. Gottheil, [1906.

partial office for a time under his uncle Mohammed, was
appointed cadi by the all-powerful Barjawan. His diploma
read that he was appointed to act for Cairo, Fostat, Alexandria,
Syria, Arabia, and the whole West.1 He had been born in
Mahdiyya in Dhulhijja, 353. He does not seem at first to have
exercised all the functions of his office, a certain 'Umar ibn
'Abd al-'Aziz ibn. Muhammad ibn al-Nu'man having the author-
ity over criminal cases. If this is his cousin, the Kunya "Ibn
'Umar " is wrong ; perhaps it was some distant relative. A few
years later, he placed Al-Hiisain ibn Muhammad ibn Tahir to
be judge in Old Cairo, Malik ibn Sa'id al-Fariki in New Cairo,
and his brother, who is called simply al-Nu'man, in Alexandria.
It is not apparent why he should in this manner have delegated
to others so many of his functions.2 But he was evidently harsh
in his judgment, for in Safer 391 (1000), or 393 (1002), 3 he was
treacherously attacked and wounded in the Jami' by a Spaniard,
so that in future he had to be protected by a body guard. In
spite of this, he seems to have been well in the favor of the
caliph al-Hakim, who gave him a house near to the KhalTj
al-Hukimi.4 He was the first Fatimide cadi to be appointed
chief preacher ; in addition, he was inspector of the mint and
chief secretary. It was his own harshness of manner that
caused his downfall. A man who brought a case before him
had been wanting in due respect. At the order of the cadi the
man was bastinadoed with 1800 strokes so that he died, and
al-Hakim commenced gradually to shear him of his preroga-
tives. According to Ibrahim Ibn al-Rakik, it was his lust for
money that really caused his downfall.

1 In the diploma of Muhammad ibn al-Nu'man nothing had been said
about the Maghrib. It is, however, mentioned in that of 'AIL Of
course, a cadi could exercise his functions only over the districts men-
tioned in his diploma. See Tornauw, Le Droit Musulman, p. 243. Al-
Husain was the first to have the title sLoJlM -Acts given to him officially.

He was also leader in prayer and surveiller of the markets. Al-Kalkash-
andi (I. c., p. 184) says : "at times the Egyptian provinces, the districts
of Syria and the lands in the Maghrib were joined together under the
jurisdiction of one cadi, and only one diploma was given him."
8 Though this was clearly within the rights of a cadi.

3 According to al-Musabbihi. In Safar 13 he is reported to have said
the prayers over the body of the vizier Ja'far ibn al-Furat. Ibn Khalli-
kan, vol. i, p. 321.

4 Khifat, vol. i, p. 71.

Vol. xxvii.J A Distinguished Family of Fatimide Cadis. 235

In Ra jab 393 (1003), though al-Husain was confirmed in his
office, his cousin 'Abd al-'Aziz was permitted to take testimony
and to act as referee. This divided authority occasioned much
difficulty for litigants. His continued haughtiness and harsh-
ness eventually robbed him of the caliph's confidence, and he
was finally removed from office on Ramadan 16, 394 (July 7,
1004). On Muharrani 6, 394, he was imprisoned by the order of
al-Hakim, and, together with two others, was beheaded at the
beginning of 395. The bodies of all three were then burned.

The place of al-Husain was taken by Abu al-Kasim 'Abd
al-'Aziz, who was invested cadi Ramadan the 16, 394. ' He had
already held office under or together with his cousin al-Husain.
It is expressly stated that he combined in his person all the
offices of cadi, even that of hearing complaints. " He seems to

1 Born Rabi' 1, 354 (=March 7, 965) or 355. The authorities disagree.
5 (VJlfe+Jt -3 .Jo-A.!! . This office seems to have been in some man-

ner similar to our court of appeal. Cases were also brought before the
official holder of the appointment which could not, for one reason or
another, be settled in the ordinary procedure. " dllix torts, est un

terme technique designant les actes injustes et dommageables que lesmag-
istrats ordinaires se trouvent impuissants a reprimer, et dont on demande
la reparation en s'adressant directement au Prince." — Leon Ostroog, El
Ahkdm es-Soulthdniye, Paris 1901, p. 209, uote. Technically, again, the
functions belonged to the vizier, though he might delegate them to a
representative. Mawardi (I. c., p. 39, 3) says in this respect : ^1 \«-sU
Lg^o v._ >.A.AA**O^ [Vu fc>»W _3 JO.AJ . Al-Makrizi is more precise ;
according to him the vizier sat as appeal judge in case he was a mili-
tary man. Khitat, vol. i, p. 403, in the chapter headed : JaxJI

A . Generally, however, the head cadi was

(5 ' '

appointed by the caliph to hear such appeals, or an officer was especially
designated for the purpose (see, also, ibid. vol. ii, p. 207). Ibn Khaldun,
Mukaddamdt, p. 193 : *5lftJlJ

|JLiiJf

See also, de Sacy, I c., notes, vol. i, p. 132; ibn Khallikan,
vol. i, p. 346, note 14.

236 72. J. H. Gottheil, [1906.

have enjoyed unwonted honors at the hands of al-Hakim, being
permitted on two occasions to enter the pulpit with the caliph —
an honor, which, as we have seen, had also been accorded to his
father. As a judge he was severe and firm. He is especially
noted for his learning in the Canon law according to the Imami
rite. He was appointed to be the head of the Dar al-^Ilm,
where he collected a large library ; he had the care of the mosques
and of the pious foundations, (wakf] , and was administrator for
various estates. The marriage of his sons to the daughters of
the Ka'id Fadl ibn Salih was celebrated in the Kasr itself; but al-
Hakim must in his madness have found some fault with him and he
was deposed on Friday, Rajab 16, 398 (March 27, 1008). That he
was married to a daughter of Jauhar has already been related.
It was perhaps this fact that hastened his fall. On Sha'ban 7,
398, he and al-Hiisain, son of Jauhar, the general in chief of the
army, were ordered by al-Hakim to remain in their houses and
not to show themselves in the market-places. This order was
rescinded a few days afterwards, and 'Abd al-'Aziz was in office
again in Safar 19, 400. l But al-Husain and 'Abd al-AzIz evi-
dently did not feel themselves safe, and together with the wife
of 'Abd al-'Aziz they fled from Cairo. They were lured back
by the crafty prince, Muharram 4, 401 (1009), 2 and the chief
executioner, Rashid al-Hakiki, with ten Turks was ordered to
put them to death. Their heads were then brought to al-Hakim.
This occurred in Jumada 2, 401 (Jan. 31, 1011).

Curiously enough, Ibn Khallikan asserts that with the death
of 'Abd al-'Aziz "the office of cadi passed out of the family of
al-Nu'man ;" but one more cadi wTas to come from the family —
the son of 'Abd al-'Aziz, Abu Muhammad al-Kasim. Perhaps
he is not mentioned by Ibn Khallikan since the exact date of his
death is unknown. Yet this would not excuse his distinct
statement in regard to the passing of the cadiship.

After the execution of 'Abd al-'Aziz, Malik ibn Sa'd al-Far-
ikl occupied the position. He remained in office until RabI' 2,
405 (1011), when the functions were given to Abu al-' Abbas

1 De Sacy, Les Druses, p. cccxxxii, says in 399.

2 Al-Makrizi in de Sacy, I. c., vol. i, p. 61. See, also, Ibn Khallikan,
vol. i, pp. 253, 345. Even the position of .vJLih-tJt -i JcUJt was again
given to him.

Vol. xivii.] A Distinguished Family of Fatimide Cadis. 237

Ahmad ibn Muhammad ibn 'Abdallah ibn Abi-l-'Awwam, who
held office until his death in Rabi', 418 (1027). Al-Siyfiti
affirms that al-Kasim was appointed immediately upon the
death of Abu-al-'Awwam ; but Ibn Hajar dates his first appoint-
ment from Jumada 1. There may, therefore, have been
another interregnum. In spite of the high-sounding titles
attached to his name in the diploma, he lasted only a little over
a year, ' being succeeded by the son of Malik, 'Abd-al-Hakim
ibn Sa'id. 'Abd-al-Hakim was in turn deposed in Dhulka'da
427 (1036) ;3 and our Kasim again returned to power, having
jurisdiction over both civil and criminal cases, and being at the
same time chief preacher. He had as assistant the historian Abu
'Abdallah Muhammad ibn Salama al-Kuda'i.8 This second term
of al-Kasim lasted thirteen years, one month, and four days
until Muharram, 441 (1049) ; but the historians do not speak
with admiration of the manner in which he held his office. For
short periods even he seems to have been replaced; at one time
by Yahya al-Shihabi, at another by al-Kuda'i.4 He was fol-
lowed in 441 by Abu-Muhammad ibn 'All ibn 'Abd al-Rahman
al-Yazuri, who was in power for seven years and was the first
to unite in his person the offices of vizier and cadi.5 The
further history of al-Kasim is unknown, and with him the glory
seems to have departed from the family of al-Nu'man.

From the biographical data given in the texts, the following
genealogical tree may be constructed :

1 Ibn Hajar says : " one year, two months, and some days." Paris Ms.
1850 has " three months and a half."
1 Al-Siyuti has 329 !
* Brockelmann, I. c., vol. i, p. 843.

4 Ibn Taghri Bird! (Abu-1-Mahasin), in his annals for the year 436
mentions the death of the Damascus cadi Muhsin ibn Muhammad ibn al-
Abbas, who was the nd'ib of al-Kasim ibn al-Nu'man for that city. Dr.
Wm. Popper, who is preparing an edition of the latter part of Ibn Taghri
Birdl's work, has been kind enough to give me this information.

5 See al-Ghumri, below. The name occurs quite often in Mss. mis-
spelled ^b .

238 7?. J. H. Gottheil, [1906.

Hayyun

i
Ahmad

I
Man§ur

Muhammad

I
Abu Hanifa al-Nu'man

Abu al-Husain 'All Abu 'Abdallah Muhammad

I I

Abu 'Abdallah al-Husain Abu al-Kasim 'Abd al-'Aziz

m. daughter of Jauhar al-Ka'id
Abu Muhammad al-Kasim
Muhammad

III. THE ARABIC TEXT OF IBN HAJAR FROM THE PARIS Ms.

No. 2149.

( ol. 136b.

£ auAwJ '

> JLjf

Lo J^!^ xJLs. J^fc jcLkJ! U! y*J! ^

-*A ilAX) cjJlLCs 8s^^X« , $.ci *-J

's ^XsxJ! --»J UolyccwLs »j>^! Lg-c.U I_j
i b

_xxxi:^U JuaJ't ,v

1 Read oL jL»Jt , and see Yakut, vol. iv, p. 381.
5 Ms. Paris 5893 has the better reading I _g ...V^ .

Vol. xxvii.] A Distinguished Family of Fatimide Cadis. 239

iaAJ JJ J o^Ls UaJ! U JLf 3

,jj
Lo

c

*AAi ^x^j [2 j>UoJI ^ UjJL? l^-

7^7*^ ^^^ u M « (; f^jj ,j.A

J! o^U-Lxs J"<3 L+_J .r^Lb -jl oL^i! JL& 5
l^ [ ? ? ] £, &jUj oJ^ . iLudJi

•no

s _xi w *_^ . «x> 15

»j sfto j»J-;. ^AJU*/ JLww _R-^

1 Mss. Paris 5893, 2152. * Mss. Paris 5898, 2152.

3 Read

240 72. J. H. Gottheil, [1806.

_?t sjJ^

LI O

x

<XJl xjulJt x2L»JI

10

fol. 85a.

^^ai 1UJ!

-Ax3 ..vX LX*L~S. VAAAA-U silJ3« ,jL«.JlAjl .>J

15 xxXi*.

u.

-XC^t '

Reading of Ms. Paris 2152, Ms. Berlin 9819. Ms. Paris 5893 has

; Ms. 2149 R^U4>! tXr^ .
9 Read jaJ! as Ms. Berlin 9819.

Vol. xxvii.] A Distinguished Family of Fatimide Cadis. 241
-A

>*>ob
'« «Xll- ylAJO «A4^ww cx& .^JL^ ti.^j ^»^>

(»LxftJ^ JULo^l^ abUarLl^ LgJUx.!^ -«3jo j^ _ a

..^* 5 Ji> J! OvAOJt jvS' . JuoUC*J!^ jjJ

1 *AX> J! S.Li! dlJ5 ^,1 ov*j viLJ! Ljt^a. !»>» viL*i

v^^' 1*^' (J^U-Aj • &*Xi (v^*^ tXA^0^5. (^-* >«Ai*lj 10
xJ-u/ aoJo ,~^^ s3*XxN °^' si' ^j

«JJ! ^y"

iJ JUi .JoUoJ' g>\ ^UJf ^.A JU^ O^ciJ! v.J-b p 15
JLlii aJLs*. ,^-c yo tXxjUA. J^.g^JI &

bf JUJ *JLo *x>L^I si 0*7^' (J

jJLJf Jjel (jd*j (S*&) |»X^! ^t *kij^ «>^^Jf sJL-ioi

242 Jt. J. H. Gottheil, [1900.

xJU* J^c jvX^I 3 Ja.L> ^b CjuSyi xJ

J! Lo Jli'j . ^AA/ob -jeliaJ! bt viU

j! «AJUO! Ljlf^ . Jmj J^y*^^ »**Lfr

!<X»»s? sis*!

*-wu
15

*-^» £ ,**

JJ^ Ij iJCyiJI

v^u ». .

!. &;;A^SX^S -sl^

MSB. Paris 5893, 2152. s Mss. Paris 5893, 2152.

Vol. xxvii.] A Distinguished Family of Fatimide Cadis. 243
AJ Lo

Jjow ^ 4^ JUJAJ ^1 ^l^ 0! Jf ^tf| w j!

. viJUjo J! iuJLkx jou jt LSb jJbb

. ajJU. J! ^UxxJI ^J "^c ^Lxi /Jl 10
^ ,3, ^jfe' &Jil iba^Jt sLjJL?

fol. 129a.

5.jcLftJ!] JLSJ ^il^wJiJI ^jJLj! ^^JLa. ,j.J
aLjLtJiXi'. ..w-x-X-jJ x l ^. JLO c^w_JL_j ^j jj. k_x._jl^_'l 15
*x bcXJL au^v^fl s^Lc.1! | J *Jkji. v_>JtJb

1 Read .j*. as in Ms. Paris 5893. s Mss. Paris 5893, 2152.

» Read yjtjf . 4 Ms. Paris 5893 sSL^ .

244 72. J. H. Gottheil, [1906.

_A^. o

<XJb (%-U ^olrf-t J^Jo . Xi ool^iUjtJ Jju ^c ilo ^ auo.j

,

>;^ * iuJI

10

^*r ^^- '-r.'*-^' ' <X

!; s

j &JLJI

15 .

.•^j xx-b*

' Delete; not in MSB. Paris 5892, 2152. s Mas. Paris 5893, 2152.

1 Read

Vol. xxvii.] A Distinguished Family of Fatimide Cadis. 245

JU . U!

li aoUj (JAIL? «jL<aJt »x) y-JLai ^1^ iUi3

. *o *x o

!j sjjy JU» ^ b! yL«J! ^^LuoJI j

4X^2 Jjct sw*4J! JLs

-XA~». ij

«iko

Jo> ,vJ . dUtX-J /-x>^-' *Uu^. cJ^ .^^oLftJ! v^JLwo LgJ ajti J ^x 10

,Jt *=>-.->' U-U . Ji '-^^ r&l ^v*^ UotXs^J I g A H

JLiii oAflJW .X *x> ^x^rv ^ft _x t/«-*

i" U-Ls aJL

cxj *X^ 1.2.3 f o^xiuixi &«MA^ o^>-vJ L^Joi> ^LiiJ!

cjL4AxJI lSxJ jLo^LxJ' ^Ax^^AJ^^t /^r1' o^; ^^ 15

1 Read ^o!il! as Ms. Paris 5893.
* Ms. Paris 5893 has AAJU!! : 2152 «n*jt .

» Read ~ <gv ... t M . * Ms. Paris 5893 L^.^ XAx

246 K. J. H. Gottheil, [1906.

OvJU

-? Jo j.Laj . JU
«JL?^ . JL? -A£

J,f Ufiyxl ,«J>5 -^oiiJt J,l oU-fjs? oJiJL? L^i! bCxL

5 ^~W«AJ ^oLoJ! (J! j»t\AJ ;JLj -Ai ciJtX^^j

JI r?\j .^s^j «-

^ _,I5UJ! ItX^o ^^

x« «>« t- v« ** JJIAJ . 'V (^ -"

JLo Ia.c,s>

V* .

x &*_N!

\j6.
.% auJ! «i^ . J^xJi ^ XA.J..XAJ ^'1

auJ! 6.3Ls

1 MB. Paris 5893 +

Vol. xxvii.] A Distinguished Family of Fatimide Cadis. 247
j! a,! jU JLL^ &A

a

LftJ! silJ3 iixi «xl^

uxxi o-X^ . JLI J^ |V^xxi xj^Jb o^So. |% g ^ « J

10

. -**+* +

i Lxi

o auo «jo L«J Lft^VwX) -iXJ3 «x ^^j . otv*Jb (joo ..^^ JU3 15
cXx£ «j! J JLj x-o. ^i

Read v*$ as Ms. Paris 5893. J Ms. Paris 5893.

» Read O« .

248 Jt. J. H. Gottheil,

a j»L*JLJ^

4JI -SOA

JU

*AA» AW. ^tJ Ntx auxi^o LJ|

^-»v-^_ ^

^jl^j . LJX jj^o ab^lc. sLx>! ^^o >s>. s^ *S^N AJ
10 .AXAJ! RA

\ o-Lt^ JoU! ^.X^jdf aJ Jco^. CiJ^ JUs .

xiLJ j&! (jl^ >sxJ«,AJ!^ (jj« JiAJl

N!J ^ J«^N5^ *

JLi' . ^tfc^^j aA »xi

/• — *
!«:>< K . ._>f 13! auJL1 *J 161

«j| Jt U^LiJ! X^AJ ^ sLaJi

Read

Vol. xxvii.J A Distinguished Family of Fatimide Cadis, 249

JLo

00^5 . kil^JUl Jl Jju JJLi ^ 1^x1$

^UJI L?t auSfc' JU 5^) J^c

«itXj . ^Jt

v_

..^JCAwli jv^.*A^=» s^^SVj '[ill] ^SAJ ^ O^xcJl ,j.x) autju! Kx

Jl_X>

K Lx) 1X4^00 jLJLi «AxxiU do iu k>^ 15

A*.J(.**2 «*wO
J-0

1 Ms. Paris 5893.
Vol. xxvii. 17

250 /?. J. 11. Goltheil, [1906.

J3\ cX^jJ jj cyL

5 jjjo LAAAJ jj^yuLiJ jJ^ R

. s ^oft c

fol. 40a

^o xLo ^j! ^1 j~°j- ^A a >^

«x)L4-Lf ^jU.xJLJ! ^.o tX^^Xx1 c\4.^ xAA^X^ul.

10 J$L£tXX**/L[i] LX>«.J ,xLt w ^ x. -.

LoJiJ! sY^ [* -x>

wi ^i J Jua -i^! ^ dJ^ ^;U*AJ! ^ <XI -

^xjL»J^ «**J iLLw J^V >«^

1 Mas. Paris 5893, 2152. 2 Mss. Paris 5893, 2152 juu»o .

3 Ms. Paris 2152.

Vol. xxvii.] A Distinguished Family of Fatimide < '<!</;*. 251

|%.jLs jio viJU

^XJU gcU

*t**jM*& 5

JLo JIAJ

r*

^,

J-SktXJ -^axJ! s^Lo SJ^.AaJI O^*AJ! xiuJ! juJLt lyi'j ,^a-fj *x!Ls.l 10
» >"/•(_

*^ •• y -^

dU6 ,

*.

s ^^.Aw- ^ . -,

CJ^LA. J, JLJ j

JU a^-U^ au

15

1 Ms.

* On margin ^L»JijJ! »L^.t ; in Ms. 2152 as a correction in the
author's hand. :l Ms. Paris 5898.

4 Ms. Paris 5893 vLoL»Jf . 5 Ms. Paris 5893

252 72. J. H. Gottheil, [1906.

. JJJ &*<

aol&Uai'! xJ'^L

i
|*J*»u)JLJ x^ftv-t Jl^xit jj-o (joZxXj ^

'. ».AJ

Ss'Lcii

10

V £?b.5 r^ ^

v^VxJ' k\A£ X»Jt , .vJ' 1<=.^aLJiJ' ~
y*y w • lj /

_XjLi . ^o^JiJ! ^AA.*W ^^-^ xA5 'iUc)' ^3 s_?*-«fl.

eu/x

Ms. Paris 2152. 2 Ms. Paris 5893.

Vol. xxvii.] A Distinguished Family of frit! m !<!<•. r,/,/,\. 253

ftJLa. Lo

*AAJ yj

o'KJ! J-o^-j . SJ>L\j

wil ^.x J^l yc^ . Ovx;.^ -«d^\.j Lo (j»ioA^i.j 5
^y> dlb Jus Jt^x>V! oJ^j . Ul

-*oxi

»LdS

v«3 v_a xj»-o o..x«.

i5l L JQ ^i -S? xJi

^..i «4-w«..) ^

isx,} fj>*i/ xJ

j ,j!

10

8*-o \jjQji* <XLJi Jv^c! Ji5l ^ ^ -

xJ'U Jui3 Lxi .£. oLoJ! »tXJ • x+JLb x; ^^t. »J LejJI

Ms. w .

254 R. J. H. Gottheil, [1906.

/-*'

5 ^ uS [foh 40b]

10 a

OwA£ ScSCkx -AJlXw

9-X2J |V^L*.! -X>! _^ ^j! J|

_j. V^V*-^ cXx£ x^fr ..vjl ^)L?

t> -s «.^-S? v..

1 Ms. Paris 5893 sJuo .

9 Ms. Paris 5893 for CI^AX.)! _3 has

Vol. xxvii.] A Distinguished Family of Fatvnide Cadis. 255

oi

i Lo JUj

J! xlo VUJ ^ ^1^ UU . ;LJL?

jJ! J^ JU» U?

1 Ux. ^IS', .^yJiJ! VL,

tf UT ^Ui iiociJ! x jJ!

Jl^ol ,j^ xlxJf auJ^ is,.xo. sjo^ iO^£ Soo Jo

j /-^i^ ioo^lxx stX^ L£XX> x-

. Li

> (_> •• o

• t !• * i n ti • •• • • i ti \ i • i •• it

2Ol . (7 r^ ? / V ^^ , " W wO . £• ^yj , tXftA A j IA<^ *^w L ^ f wVw^^I 3 *^LJiJi

^ o^ • (5' • ^ • 25 (-^' ' (.5*^ O^°^

*L\£ ^.X) &AJLC J^-4^ L-Oj SL5l &X)»..> ^ ^^^ S T^*" 2 **J)

1 Ms. Paris 5893 jCi I.

256 7?. J. II. Gottheil, [1906.

Lo *JI *3jo J^ Jb

Jt «j !o»
S. LoJiJt &*i

[ blank ] ^>J oa

10

. »>j, ^

J*& ,j*-ow^ Q^^-i.! ,»fcj -3 L
_3 JiAJ! aoJ

c^« .
_i xi'lyotj (£j£>» J^-^ ^^J (^"^^ -^XA^4JI JU . LX-^J ,

AA'jtJ! *<

fol. 73a.

Ms. Paris 5893 ^IxxJ! . s Ms. Paris 5893

3 Ms. Paris 5893

o y

Vol. xxvii.] A Distinguished Family of Fatimide Cadis. 257

lx> .x^^l &4.fc %.i*S JJ!

uli ^5 JiJ!
sLoJiJI U &JU JL^V -i aJ v.^ 5

>^o x..3 -A»O

o^xijl

Nliaj wi^ JJ5 au!

»-o
-J!

1 Ms. Paris 5893, 215.3. At this point there follows, in the Paris Ms.
2149, a portion of Ibn Hajar's biography of 'Abd ar-Rahman ibn 'Umar
al-Bulkini, f 824 A.H., which has been accidentally misplaced.

* The following up to p. 260, line 13, is taken from Ms. Paris 5893.

VOL. XXVII. 18

258 H. J. IT. Gottheil, [1906

i^ ^LJI 3 ^\S UJU

2 X+A cXx£ sjlaJ! v_A.Jlic £ !yt*« Ijjiy Lo
ii viJLJj liJLsJ &J

^ ^.x Lg.Ai

. j ,*

|*LC&.Yt ^ 5 Jo cjJouo^

>fi. ,*fcxi. ,.>JO k^LA. *^ p

MM

uo! -II &AX) 4 (vJ°^y«t ^ ^ J J.JCJ

1 Above the word V , referring to the marginal note P . Does this
mean " delete ! " ?

4 Ms. Paris 2152 * tX . 3 Ms. Paris 5893 S

4 Ms. Paris 2152 auot

Vol. xxvii.] A. Distinguished Family of Fatimide Cadis. 259

(j-i j £

*A+2»- |V-L*"J3 • &-LXj <>JW ^L«.»jJI ^J JLfc ^.J (JJ

x. >
*«_) ^ 8r*^ XAxliL! VI f-^xi^ .c^'T3. ^ *^W »t>W ^

OOOj . C J^wgJ! ^ )*7+-'' 1^"* U*''-*^^ ^LwwwJI /**4J (jl !%«-»i

! (5-Lc.

•vf Li! ^1 ^JU<3 tyii^3 . aL^LJ! wuuo! ^1 (W^Jo! XiLiJ! 10

AXJgoV

^j j y w w y w r

^ k • /

LuiLo SV^LJL'

. Je=*.L^«J! 3 JaJLJU

1 Ms. Paris 5893 , .-.v^! . * Ib. auLJJ . * Ib.

260 R. J. H. Gottheil, [1906.

dLf jJ ^* *» *

^ v
Jt sxAJ dlte tX^ju &J

j . JL»J! OAJ ..vx)

&JJ..A-1-4..M ^LAA]! ^X) Rjiios ..

; g «» ^ **••

. v^*^.. AXuw &JUw 0,O^

15

1 Ms. Paris 2152 ^X*1AA> U^. JU *l^Xi ; 5893 L^xJLft .
8 Ms. Paris 5893 Lg.o Jo . 3 Ms. Paris 2152 ^ .

* Ms. Paris 5893 &AJUO .

s From here on the text is from Ms. Paris 2149, compared with 5893
and 2152.

• The reading of 5893 and 2152; Ms. 2149 has

' Ms. 2149

Vol. xxvii.] A Distinguished Family of Fatimide Cadis. 261

UJI Jl 8

JUi

^tXJ «J ^Ls

t Lo^UaJLs

,jj v

JU 5
iL^UJf ^ x^^o Lo v*J! Jo » JoU! o

UJ! USU . !.

? 10
aUUb «J
auUCc

*Lis &AjLs»> -ff *Lx

o |^Xa ^^J' ^'o
>JtJ Li!v3 -£joCu<U «>L*JI ^ au^S' Ovio ^*JLi 15

JoUJ! rj.! ,JUU

1 Ib. > 8 Mss. 2149, 5893 ^uu*j . Ms. Paris 5893

« Ibid, auxi! .

262 It. J. H. Gottheil, [1906.

L!

wo *-• >

>g -^ ,j

t.>_*2_sxi H«Jo

*

Jo %Jc X'^aJI 2uJi

5 .

»

LM J! '^JLo^ Jo ^5^^'; -^>LoJ^ JoUJb i

^U oUaJ

_AA/

,jl . &x3. \JuLoxj -x
10

iLLu< v^1^) r^** e^0 ^— '^•^^ j^.A.»4.! *»..! ^ ^o
^ -^LoJ! *-M tXxfr ^ ^UJ! o cLi

1 Ibid b . 2 Ibid vi* . 3 Ms. Paris 2152 . « Ibid

Vol. xxvii.J A Distinguished Family of Fatimide Cadis. 263

_*isxi

aLJUs.

JL^V!

»Joo

«-L=»

JU

JoUJ!

/^*«-9 8<'^ tj -Xifc^ atfjLo auo j&. /*^;5 10

JLc. *5"L^I
yxxi J! ,*>^-9

.•Li s.j^U Juu 1^ . IJ.^XAJ *J&y*U ,v^JloL^. LftJLcf Jo
UuLa*. j-?V*^l <X^£ ,5^5 ^^V^ /*^^ (J^°^ JoLftJI /~*LS>
c^cXx^ ^.-y*/ *i ^-Lc. ^Lt^».. SVA^ V^ ^*-^-*'^ oJL*.^* &A**" 15

Ms. Paris 5893 »<Xo . 2 Ibid 2>. .

3 Ms. Paris 2149

264 7?. J. H. Gottheil, [1906.

i-t &J<X>

Jyj

o *j

-^ u

--oLftJt Vr*

10 |*A»> >iXJJ> pS" l^f *Jb UJU . »ys

_J! ^^..xx^Mu

I iaj iuLo

1K,^OU^ M- eJU x«4<1 ,*.-5 ijL> 1.11 ,J! XjotX^I US

10 ^ / O O X

i jyfl c

Vol. xxvii.] A Distinguished Family of Fatimide Cadis. 265

Jlit

alftj

Jlf ftloLoVI v-^joo JU auiiJL LJU

x-o! uftAi^S e lit JJj

j I jS

** \j"? (*-»*'^>' 10
b! AXj iuuxiUlf auLJ! ^j

X^^I (^-> -i
Lj; -oJdb

ool^> 5v-Ci.£- »*^J &

1 ? L»^jo»> . * Ms. Paris 5898 >.

» Ibid oUJ!^ . 4 Ibid 5 Jj|.^ .

5 Ms. Paris 5893, 2152.

266 R. J. H. Gottheil, [1906.

«/**/

-3 eO^ ^>i .t
5 Oj-o ^jl Jl xi'cX* u^JLb xIXU »v

,j^
v.jo -xJ! uyLoil! 5b *&j& tX3^ *L.3! &XJ

V. THE AKABIC TEXT OF AHMAD IBX SA'D AL-GHUMKI FROM
THE PAKIS Ms. No. 1850, fol. 201b.

10 •

^ -3

.4.£=L,^.

15

oj _J ^-. tl t ti * LgJ^I sLdJiJt ^o L*jJ / *n>?

Vol. xxvii.] A Distinguished Family of Fatimide Cadis. 267

sLdiJ! -*ox>

,5-" v.^t

M J — ^ J^_fc |*-^-Lx> ^Ij ^ox, sLdi Ldi

5

: ix

juo O i-)0«i

\Lx-i o\_i>. x^jw^^ JjCS -AOXJ -9 -jjjsxJL ^y^dJU

.a/)

sLaJLl!

JJJ (j*-<Xu< (J^t> (J^***' ^-U' 'Jx

Ujc5! JU

-

; — ^ ';^— ' &&-&; ^^- v^JLo &l 13 jv^LI <Jb I

uV ^ ^» a H Js _ & *xjU SWA. xj

268 E. J, H. Gottheil, [1906.

^xLu, aui j»Lfc ^s. cy^o p'jjr*-" tX+^oo ^-

vJj Lo^J iJjJU L$AX>

LxLjs sLdiJ! _ xi 5 Jou. LoLe. lj' .v^Ls xJ

.v^

« *^c s^cLc. cj^.xo. p* x.4-^'» ,5
»LdiJI U wUJf

jfc_J5 ^J« ,t, 4_M>...,^ f-lv-xVL iU.jJ! 2ULJ |*K^>i Oj-ti Jo

» — — j « — *>i — fr i» — » -C.

10 xi

Sxt ^»^ I — >f\ ...A. Ji

s — x_J! I — x — »._s>. e\ — 9

15 Ji\:sLo »lx2J aooJ> j ^»t-^ Jjj»! s\ JI. sLoJL)! JIl

^

^»t-^ j» s\ . so .

\/ A'' ' ' *

Vol. xxvii.] A Distinguished Family of Fatimide Cadis. 269

III. THE TEXT OF IBX HAJAR IN TRANSLATION.

[p. 238] Al-Nu'man ibn Muhammad ibn Mansur ibn Ahmad
ibn Hayyun1 al-Isina'ili al-Maghribi, called Abu Hanifa. His
genealogy has been given in the account of his son 'All. lie
came to Egypt in company with Al-Mu'izz from the Maghrib.
He acted as cadi in the army of al-Mu'izz ;" but the latter allowed
Abu Tahir to remain in his position. The first appointment given
to al-Nu'man was to pronounce judgment in the case of the
domain which Muhammad ibn 'All al-Madharfi'i had made a
wakf. This he had sold to pay a debt, and 'Umar ibn al-Hasan
al-'Abbasi had bought it. The sons of this one then sold
it to Faraj al-Hakami. Ahmad ibn Ibrahim ibn Hammad
asserted that it was wakf. The matter came before al-Khas-
ibi, who adjudged the property to be a wakf, [p. 239, 1.]
This judgment was then brought to Abu Tahir, who ratified
it ; but Faraj al-Hakami3 appealed (from the judgment) to
al-Mu'izz, who commanded al-Nu'man ibn Muhammad to
look into the matter. Al-Nu'man caused the testimony that
had been given to Abu Tahir to be laid before him according
to the contents of the book, i. e. the book containing the wakfs.
Al-Husain ibn Kahmash and 'Abdal-'Aziz ibn A 'yan testified to
the statements made before Abu Tahir and mentioned above.
Death, however, overtook al-Xu'man before the case could be
finished. He died on the . . . 4. He lived in Misr, but went
every morning from there to al-Kahira.

Abu Tahir remained in his position ; but al-Mu'izz gave him
as assistant 'All ibn al-Nu'man, who rendered judgment in the
Jami' al-'Atik5 also after the death of al-Mu'izz. Al-'Aziz gave
'All ibn al-Nu'man ibn Muhammad jurisdiction over the mint

1 Not Hayyan, as Brockelmann, 1. c., vol. i, p. 187, has.

2 It was customary for the cadi to accompany an army in order to
decide any case that might arise ; see, e. g., al-Makiizi in Blochet, Rev.
de VOr. Latin, vol. ix, p. 138.

3 The Mss. have ^L^j^J! ; I have ventured to read _* ; al-Dha-
habi, Al-Tanblh, p. 188.

4 Mss. 2149 and 5893 have here a sign that looks like the Arabic
numeral If ; but it may merely indicate a blank in the original. Ms.
2152 has, in fact, such a blank space. Ibn Khallikan says •' the first of
Rajab."

5 1. e., the mosque of 'Amr ibn al-'AV, the oldest one in

270 R. J. H. Gottheil, [1906.

and the mosque.1 Thereupon he came to the mosque and decided
cases. Abu Tahir was present in his own majlis and gave judg-
ment there as was his custom. Many notaries, lawyers and
merchants were there also who brought their cases before Abu
Tahir. These were conducted to the chief of police, who
imprisoned them. 'All ibn al-Nu'man interceded for them and
they were allowed to go free. Abu Tahir continued the custom
of sitting in the mosque ; his health was good until he took
cold and his side was lamed ; he became also too weak to move
around unless carried. One day at the beginning of Safar
366 al-'Aziz was out riding, and Abu Tahir, having been carried
[p. 240, 1] near to the Bab al-Diyafa, met him. Abu Tahir
asked al-'Aziz to give his own son Abu al-'Ala al-Tahir per-
mission to act as his substitute, on account of his own weakness.
But al-Mu'izz2 answered: " There is nothing left but to dismiss
him." Three days afterwards he deposed Abu Tahir and
appointed 'All ibn al-Xu'man, as is related in the history of
that man.

'All ibn al-Nu'man, ibn Muhammad ibn Mansur ibn Ahmad
ibn Hayyun al-Maghribl al-Kairuani al-Isma'ili, of the 4th cen-
tury, was born in Rajab 328, and came with al-Mu'izz from
al-Maghrib, and was ordered by him to preside as judge. He
and Abu Tahir acted in that capacity, the witnesses testifying
before both ; but the confrontation of the witnesses took place
before Abu Tahir.3 When al-Mu'izz died, the jurisdiction over
the mint and over the two mosques was given to 'All ibn al-Nu'-
man. He came to the Jami' al-'Atik, and held judgment there.
But Abu Tahir continued to hold court in the Jami' and he dis-
charged some men.4 However, he became partially lamed, and

1 This ought probably to be " the two mosques " — as was the case with
his son 'All, and as Ibn Khallikan says.

2 All the texts have "al-Mu'izz." It must be "al-'Aziz." De Slane
translates "to make cat's meat of." See his note to Ibn Khallikan,
vol. iii, p. 573.

3 1 imagine that this is the meaning of the phrase

4 This might be the translation of the reading in Ms. 2149. Another
possibility is the reading aLel+i! J<X&; "and to arbitrate [in] the
assembly."

Vol. xxvii.] A Distinguished Family of Fatimide Cadis. 271

al-Mu'izz turned over the (whole) office to Abu1 'All ibn al-Nu'-
man, two days before the end of Safar 366. Abu1 'All then
rode together with a large multitude to the Jami' al-Azhar, being
clothed with the robe of office.2 He was invested with the
sword, and before him were placed robes in covers to the number
of seventeen.3 [p. 241, 1] His patent was read in the jfimi' while
he stood upright. Whenever the name of al-Mu'izz was men-
tioned, or the name of one of his family, he gave a sign that all
should bow low." Then he went to the Jami' al-'Atlk in Misr
and found the preacher, 'Abd al-Samih, awaiting him in the
jami'. The time had already come for him to go out. He
then read the Friday prayers, and his brother Muhammad
published his diploma. In this, it was stated that he was
appointed cadi over Egypt and its dependencies, preacher, Imam,
inspector of gold and silver, of weights and measures. Whereon
he returned to his house, and there came to him a multitude of
notaries, claimants, merchants, and the chief men of the land.
No one kept away from him. In his patent it was said : "When
one of the parties to a suit brings a case before thee and the

1 So the Mss. ; delete.

9 The khil'a of the cadi was made of wool, without any border ; gen-
erally of white, the inside being of green (Quatremere, I. c., vol. ii, part
2, note 73). Al-Jauhar, when he came to Egypt, ordered the preachers
to wear white, evidently an anti-'Abbasid practice. See Ibn Khallikan,
vol. i, p. 844 ; JA. 1836, 3, p. 51, Abu-1-Mahasin, ii, 408. The Alid colour
was in reality green ; but the Carmathians, also, adopted white. See
de Goeje, Memoire sur les Carmathes du Bahrein, Leiden, 1886, p. 179.
These garments were part of the insignia of office, and are even called

Dozy, Dictionnaire .... des noms des v&ements, p. 14. They

were probably kept in the o^X)l ioL=»., which was a separate

department in the caliph's palace. See al-Kalkashandi, 1. c., p. 175.
Abu Yusuf al-Ansari (see p. 230, note 4) is said by Ibn Khallikan to have
been the first to give a distinctive dress to the learned (Ibn Khallikan,
vol. iv, p. 273).

3 These numerous robes had probably no further significance ; they
were nothing more than presents.

4 The text has ^J i. e. y^o^=: L*j " indicavit aliquem." See Dozy,

vol. ii, p. 844 ; de Goeje in the glossary to al-Tabari, p. DXLVIH. Al-
Hakim ordered the people to stand up at the mention of his name.
Ibn lyas, Ta'rlkh Mi$r, p. 53.

272 7?. J. H. Gottheil, [1906.

other party brings it before some one else, do thou cause the
two to come to thee." He knew that this had reference to his
withholding cases from Abu Tahir. From that day he did so
withhold cases from Abu Tahir.

Upon the third day after his appointment 'All ibn al-Nu'man
rode to the Jami' al-'Atik, with a red basket before him.1 He
took his station where the people were sitting in rows near the Hal-
kat al-Zawal.4 There came to him the notaries, the officials, the
lawyers and merchants — a very great multitude. He acted as
judge for the people, addressed the agents, and read to them
the Sura "al-'Asr" 3 impressing upon them the fear of God.
He then asked for the notaries and for the cadi, Abu Tahir.
Al-Husain ibn Kahmash, the chief notary at that time, said :
" He is still in office." Ali replied, "Let him give judgments
in his own house and not sit in the mosque." Abu Tahir heard
of this ; he discharged the lawyers and he was practically deprived
of his functions. Some people, however, interested themselves
in Abu Tahir, [p. 242, 1] and procured an edict to the effect
that he might still act as judge. The notaries were collected
and the edict read to them. This intended action coming to
the ears of Abu Tahir, he forbad it, saying: " What can I do ?
All my strength is gone." Al Husain ibn Kahmash answered
him: "May God recompense the cadi." 'All ibn al-Nu'man
desisted from demanding the diwan al-hikm; and did not ask
or demand it further — [thus showing] his good manners and
his kind actions. When Abu Tahir desisted, 'All ibn al-Nu'-
man had a free hand in his jurisdiction.

6f -
1 I do not know to what this refers. Al-Shirazi speaks of a l/-)t«>

(KauTTTpa) being placed in front of the cadi, while he is holding court (al-
Tanblh, p. 315, 10 ; cf. 321, 5, 8). Does the text here refer to this?

~ - • am at>raid that
there are some technical terms here which I do not understand. xJL=&.
may mean ''the enceinte" of a building or a gathering of students around
a professor who is teaching them, or even the hall where such teaching
is given (see Quatremere, Histoire des sultans Mamlouks, vol. i, part 2,
p. 197). JUx may mean " the afternoon." "And he took his seat in
the Majlis al-Saff at the afternoon assembly."
3 Sura 103.

Vol. xxvii.j A Distinguished Family of Fatitnide Cadis. 273

'All then appointed his brother Muhammad and al-Hasan ibn
KhalTl the Shfin'ite jurist; the latter with the special provision
that he was to deliver judgments according to the Isma'llI rite,
not according to the Shaft 'ite. This latter was accustomed
to sit in judgment when Muhammad was otherwise engaged;
for 'All had placed his brother Muhammad over Tinms, Dami-
etta, Farama, etc. Muhammad went out there and appointed
deputies ; then they returned and 'All put up a curtain in his
house.1 Now when al-'Aziz went forth in 368 to make war
upon the Carmathians,2 'All went along with him, putting
his brother Muhammad in his own place. Many people spread
the report that al-'Aziz [had superseded him]. Muhammad
thereupon wrote to his brother about the rumor. 'All at once
procured an order from al-'Aziz to the chief of police, Hasan
ibn al-Kasim, that he should look into the matter; but Hasan
answered allaying his fear,3 and reporting that the position of
Muhammad ibn al-Nu'man was gaining strength.

Formerly the notaries had been accustomed to sit in the jami',
according to the prescription of the cadis before him — in winter
in the maksura4 and in summer near to the window ; . but the

1 For what purpose this is mentioned here I do not know ; the mean-

o „

ing of '_A^ is clear. Perhaps he wished in this way to have more

privacy.

4 At first, the Carmathians were used by the Fatimids in reducing
Syria to Shl'a rule : later, the two became bitter enemies, the Car-
mathians making common cause with the Abbasids at Baghdad. See
de Goeje, Memoire, etc., pp. 133 seq. On the victory of al-'Aziz over
them in 368, see ibid., p. 1»2.

3 Reading o»i»l |»Jou ; if ,,o«J»-! , " advising him not to press the
matter."

4 The maksura was a small space partitioned off by a grille and near to
the minbar of the mosque, in which the ruling prince took his place to
say his prayers and to hear the khu^bah (Quatremere, Histoire des sul-
tans Mamlouks, vol. i, part i, p. 164 ; vol. ii, part i, p. 283). The
'Umayyad Caliph Mu'awiya preached from such a maksura (Goldziher,
Muhammedanische Studien, vol. ii, p. 41). On such a maksura in the
Azhar, see Van Berchem, 1. c., p. 47. A description of the maksura in
the mosque of 'Amr in Fos^at is given by Ibn Dukmak

sLoxVI JJic «Ja.u/! J (Cairo, 1809, A.H.), part iv, p. IA, below. Al-
Makrizi, in speaking of the chief cadi's procedure (al-Khifat, vol. i, p.

VOL. XXVII. 19

274 J?. J. If. Gottheil, [1906.

distance induced 'All to order that they should sit near to him,
[p. 243, 1] two to his right and two to his left, and that they
should thus see whatever judgment he gave. It was customary
for his scribe to charge for the documents which he signed.
But after 'All ibn al-Nu'man had been in office for a year he
grew to dislike this arrangement and forbade it. During his
tenure a man apostatized. With the permission of al-Aziz,
'All struck off his head.

Ibn al-Nu'mfin was on very friendly terms with al-'Aziz, as
his father had been with al-Mu'izz — sitting with him, eating
with him, riding out with him, and conversing intimately with
him. The vizier, Jacob ibn Killis, opposed him and the cadi tried
to ignore the vizier. This went so far that 'All could not give
any judgment, nor appoint an assessor' to the cadi nor a deputy,
without the vizier taking a hand in the matter. The cadi did
away with the attendance (of litigants) in the mosque, because
it enabled the vizier to decrease his power. When the
vizier was put under arrest, 'All ibn al-Nu'man returned to his
former custom.

He was the first to have the title chief cadi over the whole
of Egypt, because in his diploma it was stated that all the
provinces (of Egypt) were under his jurisdiction.2

Muhammad ibn al-Nu'man ibn Muhammad ibn Mansur ibn
Ahmad ibn Hashiib 3 al-Maghribi al-Kairuani the Imami, of the
fourth century, born on the third of Safar 340 [A.H.] in the

403) says : 2U<Xj ..-o .^st sXs-f ^x> au»+:a».

place m which the vizier gave

audience was divided by a grille into two parts. All who had business
with him assembled in the aLcli' ; but he sat in the maksura, which was
separated from the rest of the room by a heavy grille. This was, no
doubt, done to protect his person from sudden attacks. See Ravaisse,
.Essen" swr fhistoire et sur la typographic du Caire in the Memoires . . .
de la Mission Areheologique Frangaise du Caire, vol. i, p. 54.

1 On the functions of the JJLfr see Quatremere, I. c., vol. ii, pa rt 8, p
111 ; Ibn Khallikan, vol. ii, p. 367, note 5, and Dozy, vol. ii, p. 103.

2 Ms. Berlin 9819 adds five lines of poetry by him on the authority
of al-Musabbihi, part of which are given by Ibn Khallikan. It adds
also that he died on Rajab 6th, 374 [A. H.]

3 Of course, the proper reading here is ,*x=k in place of

Vol. xxvii.] A Distinguished Family of Fatimide Cadis. 275

Maghrib. He came to al-Kahira in company of his father with al-
Mu'izz. He represented his brother All ibn al-Nu'man towards
the end of the latter's incumbency. After the death of his
brother, al-'Aziz endowed him with full powers, [p. 244, 1] on a
Friday, seven days from the end of Rajab 374. He received inves-
titure and donned the sword. On the same day he went to Misr
in a palanquin,1 resting upon a mule because of a sickness from
which he was suffering. He entered the mosque, but was unable
to sit down; so he returned to his dwelling. However, his son,
'Abd al-'Aziz, the children of his brothers, and a large atten-
dance, remained seated in the mosque until, after the usual Friday
prayers, the diploma was read out appointing him cadi over the
whole of Egypt, Alexandria, the two sacred places, and the prov-
inces 2 of Syria. He was also clothed with the functions of lead-
ing prayer, of assaying gold and silver, and of controlling weights
and measures. In his diploma both his father and his brother
were mentioned with praise. Then he sent his nephew al-Hasan
ibn 'All- to the jami' to sit there as judge and informed the
lieutenants of the provinces3 (of this). On Friday, Jumada
1, 375, he betrothed his son 'Abd al-'Aziz to the daughter of
Jauhar, the ka'id, in the majlis of (the caliph) al-'Aziz,
The gift to the bride's parents amounted to 3,000 dinars. The
two witnesses were Muhammad ibn 'Abdallah al-'Utaki and
'Abdallah ibn Muhammad ibn Raja. al-'Aziz made presents to
the husband, and Muhammad ibn al-Nu'man went away with a
large and select assembly of his friends. His son 'Abd al-'Aziz
was appointed his substitute, his nephew al-Husain ibn 'All
beinar removed. Al-Musabbihl says that Muhammad ibn al-

o . •/

Nu'man was a good judge, well educated, [p. 245, 1] and learned
in history. Al-'Utaki,4 in his history, says while al-Mu'izz was

1 On the kubba or palanquin, see Ibn Khallikan, vol. iii, p. 846,
note 10.

2 Syria was divided into five provinces (\>LL&.|), to wit: Damascus,

Emesa, Kinnesrin, Jordan, Palestine. See Lane, p. 470a.

3 _=J«jJ! oLftJLs* . On the meaning of Khalifa (lieutenant), see the

article of de Goeje referred to in Van Berchem, I. c., p. 755. Al-Shlrazi
(al-Tanbih, p. PIP) speaks of the Khalifa of a cadi.

4 1 have adopted the reading of Ms. 5893, and believe that the refer-
rence is to Abu 'Abd al-Rahman Muhammad ibn 'Abdallah al-'Utaki,
author of a Ta'rikh al-Maghariba; Al-Dhahabi, Mushtabih, p. 84b. The
title alone is mentioned in Ibn al-Faradi, Kitab 'Ulama al-Andalus
(Madrid 1891-2), p. 811.

276 R. J. H. Gottheil, [1906.

still in the Maghrib he ordered the cadi of his land, al-Nu'man.
ibn Muhammad, to make some silver astrolabes, (advising him) to
seat some trustworthy person near the workman. Al-Nu'man
placed there his own son Muhammad. When the work \\;is
finished, he gave it to al-Mu'izz, who asked him, "Whom didst
thou place by the side of the workman?" He answered, " My
son Muhammad." Upon this the caliph said, "He shall be the
cadi of Egypt." Muhammad ibn al-Nu'man related: "When
al-Mu'izz saw me — I was then quite a young fellow — he said to
his son, ' Here is thy cadi."' Al-Musabbihi relates : "Muham-
mad ibn al-Nu'man during his term of office appointed thirty
assessors."1 He says further: "Muhammad ibn al-Nu'man was
very subtle in his judgments. A woman once came to him
demanding her rights of her husband, which he refused to render
to her. She then asked the cadi to put her husband under lock
and key, which he ordered to be done. Looking at her, he
found her to be comely and in a gleeful mood. When her hus-
band went to the prison, the cadi ordered that she should be
locked np with him. At this she was very angry ; but the
cadi said to her : ' We have locked him up to satisfy your rights ;
we lock you up to satisfy his.' When the woman found this
out she withdrew the complaint, and as she went away the cadi
said : ' I saw that she was delighted at his being locked up, and
I was afraid that she would have leisure for wrong-doing."1

He adds : The vizier ibn Killis was much opposed to the sons of
al-Nu'man in regard to their judgments. It happened that al-
Hasan ibn al-Husain ibn 'All ibn Yahya al-Dakkak married | p.
246, 1] his son to an orphan known by the name Bint al-DibajI
with the permission of Muhammad ibn al-Nu'man.2 One of the
witnesses, Bakr ibn Ahmad al-Maliki, arose and charged that the
contract was fraudulent, as the girl was not of age. He was per-
sistent in this statement. Ibn al-Nu'man, however, decided "It
has been proven by her own assertion that she is of age. " She was
taken to the kasr and the affair was brouht to al-'Aziz. She

1 I suppose that this is the meaning of JtXt here. It might also sig-
nify " he reconciled," i. e., litigants, which was one of the functions of
a cadi.

- Wards in chancery could be married only with the permission of a
judge. Al-ShirazI, 1. c., p. 19.

Vol. xxvii.] A Distinguished Family of Fatimide Cadis. 277

was examined and found not to be of age. The cadi was there-
fore asked to annul the marriage ; whereon the vizier cited both
cadi and witnesses before him and having sworn them, said :
"May his honor declare this marriage annulled and refuse to
receive the testimony of these witnesses." The cadi did so,
putting out a document to this effect. In it the statement was
made that it had been proven to him that the girl was not of
age. The vizier disapproved of the witnesses as they had been
negligent in their conduct. This happened towards the end of
Jumada 1, 375. He (the vizier?) commanded that the property
of the young woman should be placed in chancery, though he
deducted from it a quarter for her maintenance.

(On another occasion,) it was reported to Muhammad ibn
al-Nu'man that a Christian had tui-ned Mohammedan, that he
had changed back again, though he had passed his 80th year.
He was asked to recant, but refused. His case was reported to
al-'Aziz, who had him turned over to the chief of police. He
then ordered the cadi to send him four witnesses who should
induce him to repent. Should he so repent he was to have (from
al-'Aziz) 100 dinars ; but if he persisted, he was to be killed.
He rejected Islam and was killed, his body being cast into the
Nile.1'

A man of the Walad 'Akil ibn Abi Talib2 once cited before
him his wife who had a daughter with her, of whom the man
declared he was not the father. Ibn al-Nu'man tried to dis-
suade him, but was unable to do so. [p. 247, 1] The case was
brought before al-'Aziz, who commanded the cadi to have the

1 On the treatment of apostates from Islam, as recommended by the
various schools, see Goldziher, Muhammedanische Studien, vol. ii, p.
215; Tornauw, 1. c., p. 298. Al-Hakim, of course, handled with great
severity those who were enemies of the Alid pretensions. It is related
that a Syrian once affirmed that he did not know who 'All was.
Muhammad ibn al-Nu'man did the bidding of his master, had the man
imprisoned and then sent four notaries to question him. After that, he
w*as brought before al-Hakim, who had his head cut off. See de Sacy,
Les Druses, vol. i, p. ccxcvm.

2 'Akil was a cousin of the prophet. See al-Nawawi, y^j JL$J wUc^
£>L»»u;^l! ed. Wiistenfeld, p. 426 ; Wiistenfeld, Register zu den genealo-
gischen Tabellen, p. 84 ; Sprenger, Mohammad, vol. i, p. 146 ; al-Dha-
habi, al-Mushtabih, p. 368.

278 ./?. J. H. Gottheil, [1906.

li'an1 pronounced between them. The cadi cited them in Dhu-
1-Ka'da 378 to the Jami' al-'Atik; the witnesses came and he
warned the husband, who, however, insisted upon the li'an.1
So the cadi pronounced it between them and thus separated them.

His son 'Abd al-'Aziz was appointed judge in his place.2 He
presided on Mondays and Thursdays. At the beginning of the
year 381 he appointed many of the nobles his assessors ; and in
Safar 382 he appointed a man named Ja'far in the jami' to
deliver fetwas according to the Meccan rite. But the faklhs
of the jami' rose up in tumult against him. When the cadi
heard of this, he took some of them and sent three of them
around (the city) riding upon camels.3 As the position of the cadi,
Abd al-'Aziz, became more secure, he ceased altogether to go to
the jami', holding court in his own dwelling place. No one
spoke to him without addressing him as " Our lord ".

Now when (the caliph) al-'Aziz died, Muhammad ibn al-Xu'-
man remained in his house in al-Kahira, and had his son 'Abd
al-'Aziz preside at the court in Misr every Monday and Thurs-
day. Ibn Zulak says : I have never seen of any cadi in Egypt
what I have seen of Muhammad ibn al-Nu'man, nor have I
heard the like of it in regard to any cadi in 'Irak. He deserved
this reputation, for he was learned, careful and cautious, distin-
guished in bearing and in conduct. Abu 'Abdallah al-Samar-
kandi4 says of him:

1 The li'an is the curse which the husband pronounces upon his wife
if he suspects her of adultery, but has not sufficient proof to substan-
tiate his charge, or if the fourth of the necessary witnesses is wanting.
The formula that he uses is from the Koran, Sura xxiv, 4-9 &JJI JUxJ

.wOi^lSsJ! .wo o-*^ ,.)? LgjJLc. , upon which the whole ordinance is
based. The wife could do the same in regard to her husband. See the
section ^LxJLjf s^La in al-Shirazi, I. c., pp. 233 et seq.; al-Sha'ram>
Kitab al-Mlzan, ii, 111 ; Kashf al-Ohumma, ii, 86. A portion of the pas-
sage in Bukharf s Sahih is translated by Goldziher in his Muhamme-
danische Studien, vol. ii, p. 235. See, further, Tornauw, 1. c., p. 219;
Querry, Droit M usulman, vol. i, p. 92, and Snouck-Hurgronje in ZDMG.,
vol. liii, p. 163.

2 Ibn Khallikan (vol. ii, p. 365) mentions the fact that in Jumada 1 , 380
(July- August 990) Muhammad appointed the celebrated astronomer 'AH
ibn Yunus, the author of so-called Hdkimite Tables, to act as 'adl.

3 1 suppose that this was considered degrading for a man of position.
Only the poor and the Bedouin ride on camels.

4 Ibn Kallikan has "Abd Allah ibn al-Hasanal-Ja'fariof Samarcand;"
he, also, cites some more verses than ibn Hajar.

Vol. xxvii.] A Distinguished Family of Fatimide Cadis. 279

[p. 248, 1] He was unique in noble qualities, he was illustrious

in honorable deeds and excellent.
His brilliancy gleamed and he pressed resolutely on, as gleams

a polished sword.
When he,gave judgment rectitude was his companion, when, he-

gave awards beneficence was his colleague.1
When he ascended the pulpit he was a veritable Kuss, when he

was present at gatherings he was a true Khalil .*
Al-Musabbihl says he wrote many verses, but they do not
evidence much power. Among the best are :
O thou who are like the full moon of heaven, when seven and

five and two days have passed;3
O thou who art by nature perfect in beauty, thou engrossest my

heart and keepest my eye from sleep.
Is there anything that is desirable in thy mouth for me ?4 If not,

I must go off with the sandals of Hunain.5
He adds : During his term of office he had a brutal fellow
stoned who had committed adultery with an Alid woman.
The man was stoned in the Suk al-Dawabb near to the mosque
of Ibn Till fin in the year 392. But when increasing power
came to him and his station became elevated, sickness took hold
of him — gout and festering sores. He was ill most of the time,

1 Some readings in the Mss. are undoubtedly wrong ; Berlin 9819 omits
the lines altogether. In the first line of poetry, Paris 5893 has x

Ibn Khali. aJjdJ for sLstflJ' . For *!A« , Ibn Khali, has tXx=»* .
In the second line for Lol-JCct , Ibn Khali, has Loyx£.| ; in the third,
Paris 5893 and Ibn Khali. 4>tjk.**JL for t>f**uJL . I have translated
according to the emended text.

2 According to de Slane(Ibn Khallikan, vol. iii, p. 573/4) the references
here are to Kuss bishop of Najran, and to al-Khalil ibn Ahmad. Kuss
was renowned for his eloquence, and the saying went : ^x _ . U j^t

,ujj' (see Freytag, Proverbien, vol. iv, Index) ; Sprenger, Mohammad,
vol. i, p, 102 ; Sprenger, El-Mas'udVs . . . "Meadows of Gold," i, p. 138.
Al-Khalil was the founder of the science of prosody among the Arabs
(Flugel, Die grammatischen Schulen der Araber, p. 87).

3 I. e., at the time of its fullness.

4 I. e., Can I expect any favour from thee?

5 1. e., disappointed. See the explanation in de Slane, /. c., vol. iii, p.
573, note 8. '

280 11. J. H. Gottkeil, [1906.

so that his son 'Abd al-'Aziz performed his judicial functions,
kept the registers in his father's house and performed his other
duties. Barjawan,1 in spite of his rank, visited him every Thurs-
day. He was very well-mannered, of fine stature, stout, well
ridden, well groomed and perfumed, whether sitting in his
majlis or riding out. Whenever he had to make a gift he gave
much and he gave quickly.

His death happened on Tuesday evening, Safar 4, [p. 249, 1]
while he was still in office. Al-Hfikim came, said the prayers over
him in his house, and had him buried beiieath its kubba. After-
wards his body was transferred to the cemetery. He had been
in office fourteen years six months and ten days. There was
found charged to him property to the amount of 36,000 dinars
belonging to orphans and others. The governor,2 Barjawan,
ordered all that was found to be seized, sending his secretary,
Abu al-'Ala Fahd, a Christian, to seize this property, to give
orders in regard to its sale, and to exact payment from the nota-
ries in whose charge the property had been. He who could
show a written document of the cadi was allowed to remain in
possession of what he had ; but he who could not show a writ-
ten document of the cadi was fined,3 until half of the judg-
ment was paid up. Then a settlement was made with the creditors
at the rate of one-half. The judge further ordered that in future
no money belonging to orphans or to persons absent should be
deposited with any notaries. He then set apart a place in the
Zukak al-Kanadll 4 where this money should be deposited. Four
notaries were to put a seal upon it, and it was not to be opened
except in the presence of all of them. For some time the matter
remained in this fashion. Muhammad ibn al-Nu'man, however,
gave over to one of the notaries, 'Abdallfih ibn Ahmad ibn
Muhammad al-Midadi, the money belonging to an orphan, and
demanded a notarial document in regard to it. The notary

1 On the eunuch Barjawan, the regent for the young caliph al-Hakim,
see ibn Khallikan, vol. i, p. 253 ; Stanley Lane-Poole, History of Egypt,
p. 124.

* The term *^l^ acquired the meaning "governor;"' then it was

used for any ruler. See Van Berchem, 1. c,, pp. 205, 420.

3 Read * *£. for *•._£. in the text.

4 Ibn Dukmak, 1. c,, p. 13, mentions a JuJUJLM OU»V > feu

/

Vol. xxvii.] A Distinguished Family of Fatlmide Cadis. 281
refused. Then Muhammad said: "He is not the one who ouo-ht

" O

to write out such a document." It happened that al-Midadl
died in the year 379, while holding many such deposits. Ya/Id
ibn al-Sanadi, the secretary of al-Hakim, had sent to him
before his death [p. 250, 1] to get a receipt in regard to the
moneys that he held. After his death, the greater part of this
money was not found. So the cadi sold his house for 5,000
dinars and with this paid the deposits.

Al-Husain ibn 'All ibn al-Nu'man ibn Muhammad ibn Mansur
ibn Ahmad ibn Hayyun (with unpointed /ta, ya, silent wan
with cl.amma, and at the end a nun), al-Maghribl al-Isma'ili of
the fourth century. He was born in Mahdiyya, two days before
the end of Dhu-1-Hijja 353. When quite young he came with
his father to al-Kahira. He learnt a book on law by heart, and
was so able. as to become one of the imams of the Seveners.1 His
uncle, Muhammad ibn al-Nu'man, appointed him judge in the
jami'. Then he was deposed in favor of the former's son, 'Abd
al-'Aziz ibn Muhammad. When Muhammad ibn al-Nu'man died,
Egypt remained without a cadi for 19 days. Then Barjawan
invited him by the order of al-Hakim and made him cadi ; plac-
ing his cousin 'Abd al-'Aziz ibn Muhammad ibn al-Nu'man over
cases in appeal. This occurred at the end of Safar or at the
beginning of Rabi' I, 389. Al-Musabbihl gives the corrected
date as Safar 23rd. He says : Barjawan clothed him with a
sword and with white linen2 garments, he put on him a mantle
and gave him a turban: — both of them gilded.3 He caused him

If this translation is correct, it will be necessary to read

On these "Seveners" see Macdonald, Development of
Muslim Theology, p. 42. Perhaps the correct reading is iuuuiJt.

2 I have translated in this fashion because of the note in Dozy, Diction-
naire des Noms des Vetements, pp. 180, 1, s. v. *$aJLo pi. m^aljLo . But

x,£.«J£JLo may be the same as the more usual jutia&o , of which de Goeje
(Glossary to Tabari, p. CDXXIX) has collected a number of instances in
the sense " vestimenta consuta."

3 Read here .wuJB joo . It seems that the turban worn by jurists
was thicker than that worn by ordinary Muhamrnadans. For that

reason, a jurist is sometimes called Rx+jL ^»\ or

(Dozy, Vetements, p. 307). The &clio or Ld-Jb (a veil) was origin-

282 E. J. H. Gottheil, [1906.

to ride upon a mule1 with two mules going before him. Before
him were borne many fine garments. The diploma, which created
him head cadi in Misr, al-Kahira, Alexandria, Syria, [p. 251, 1]
the two saci'ed places, the West and its provinces was read out
while he remained standing. He was appointed leader in prayer2
and controller of the markets.3 He rode to the jami' and he
refused to receive a company of notaries who had served his
uncle to the number of fourteen. Al-Musabbihi gives their
names. After a month's time he received them and installed
al-Husain ibn Muhammad ibn Tahir as judge in Misr and Malik
ibn Sa'id al-Fariki in al-Kahira. He made his brother al-Nu'man
inspector of weights and measures.4 This last one was also
made cadi in Alexandria and Ahmad ibn Muhammad ibn Abi

ally worn only by the chief cadi, and became thus part of his investiture
(ibid. , pp, 255, 279). Al-Nuwairi (quoted by Quatremere, L c. vol. i, part
i, p. 21) says distinctly in speaking of Malik Sa'id, son of Baibars :

xl **'&\\ _*f:Li' _JLt V! x^^JajL} . See further citations ibid.; and

O O_ }

cf. Sprenger, " Eine Skizze der Entwickelungsgesch. des muslimischen
Gesetzes,'' in Zeitschrift fur vergleichende Rechtswissenchaft, vol. x,
p. 23. Evetts, Coptic Churches, ii, p. 120. This peculiar head-dress
persisted down through Mameluke times, and is often referred to. See
the account of Barbafella, secretary of the Venetian Embassador in
1503 in Patton, Hist, of the Egypt. Revol. i, p. 62, and Joseph ibn Isaac
Sambari's Hebrew account (end of the seventeenth century) in Neu-
bauer, Mediaeval Jewish Chronicles, i, 115.

1 At a later time a special piebald mule was kept in the royal stables
for the use of the chief cadi. See al-Kalkashandi, 1. c., p. 184; al-

Makrizi, al-Khitat: JLc *j*

2 xJLo pi. c^^wO may also mean "gifts," "donations." See de Goeje
in the glossary to his edition of Baladhurl, p. 108. " Controller of gifts
and donations ?"

4 The correct reading is l^A+J! as in Ms. 5893. ^LxjiJI in the other
Mss. is a mistake for

Vol. xxvii.] A Distinguished Family of Fatimide Cadis. 283

'Awwam responsible for the pensions. He who had charge of the
property of orphans1 was also intrusted with the accounts.

On Safar 3, 391, while he was sitting in the jarai' of Misr
expounding the law, the evening prayer was offered. It had
just been commenced when a Maghrib! from Andalusia came
and attacked him. With the knife of a basket-maker he gave
him two cuts, in the face and on the head. The man was caught,
killed and hung up. From that day on al-Husain was guarded
by twenty armed men. Al-Musabbihl mentions this matter
in his history while discussing the events of Muharram 2nd,
393. The cadi waited until his wound was healed; then he went
to al-Hakim. He received investiture, was carried upon a mule,
while another was led before him.

This al-Husain was wounded while performing a rak'a [p. 252,
1] during the evening prayer. For this reason the guards were
accustomed to take their station back of him with drawrn swords
until he had finished ; then they said their prayers. Al-Musab-
bihl affirms that he was the first cadi to whom this happened.
Al-Hakim gave orders that double the salary, presents and appa-
nages2 of his uncle should be given to al-Husain. He made
a condition, however, that al-Husain should not touch even a
single dirhem of money belonging to the people.3 He put him
in office, invested him with the sword, caused him to ride on a
mule, giving him the care of justice over his whole kingdom.
He made him preacher and imam in the chief mosques ; 4 gave
him the care of them and of other mosques; appointed him
inspector of the mint, and of preaching, as well as chief reader
and chief scribe at the evening levee. He was the first 'Ubaidi
cadi to be preacher. On account of his poor health the people
imagined that he would not preside at court, and that the

1 Read -Lo^l with Ms. 5893.

2 i^jLc-Uai! either "apanages" (Quatremere, 1. c., ii, p. 200), or "the
revenues of his estate" (Dozy, vol. ii, p. 374).

3 1. e. money of orphans and the like, placed for safety in the diwan
of the cadi.
4 If the correct reading here is, as I suppose, aLxvoli! JokLwutJ! , the

reference must be to the "cathedral mosques" (if such a barbarous
term be permitted), where the Friday prayers were said. The older
term was masjid. In course of time, any mosque was called a jami'-
See the learned notes of Van Berchem in his Corpus, pp. 173, 765.

284 E. J. H. Gottheil, [1906.

appointment was really meant for the son of his uncle Muham-
mad, 'Abd al-'Aziz, because his father had preceded him as
judge and had instructed him during his lifetime. Then a num-
ber of people made the complaint that there were deposits
belonging to them in the cadi's diwan.1 The cadi cited his cousin
Abd Al-'Aziz ibn Muhammad ibn al-Nu'man, and wrote to his
uncle Abu Tahir ibn al-Musnadl in regard to the matter. He
was told that his uncle had made free use of all this money,
regarding it as a loan. Their reply displeased him, and he .made
a further investigation, sending Fahd ibn Ibrahim the Christian,
the secretary of Barjawan, .to examine into their accounts. So
he took charge and investigated and compelled 'Abd al-'Aziz
[p. 253, 1] to sell whatsoever his father had left. Then he sold
all that he had, realizing from this more than 7,000 dinars.
The secretary, however, had calculated the liability to be twice
this sum. The cadi, sitting in the kasr, called the creditors
and paid them the sums owed.

He then set apart in the Zukak al-Kanadil a special place for
the moneys deposited with the cadi and placed there five nota-
ries to register whatsoever was brought there and attested.
He was the first to set aside a separate place for the care of
moneys intrusted to the cadi. Previous to this all such property
had been intrusted to the cadi in person or to his assessor. Al-
Husain attended to all affairs with harshness and severity.2
He was the first one to have the title chief cadi in his patent ;s
his father being the first of the Egyptian cadis to be addressed
by such a title. Al-Hasan al-Maghribi once brought a law case
before him ; but his tongue slipped while he was addressing the
cadi. This angered the cadi, who sent him to the prefect of
police and he was punished in the presence of the cadi's chamber-
lain with 1800 strokes. He was carried about the city in dis-
grace and died upon that same day. His bier was brought out,
most of the people of the city came to see it, and honored his
grave, praying for him but cursing the one that had done him
injury. The cadi repented of his deed — but his repentance was
too late.

Cf. Dozy, vol. ii, p. 792.

2 auL&x> i. e. he inspired fear among people.

3 Or ''in his protocols."

Vol. xxvii.] A Distinguished Family of Fatimide Cadis. 285

In Rajab 393 al-Hakim gave 'Abd al-'Aziz ibn Muhammad
permission to try cases and to take testimony, though at the
same time he confirmed al-Husain in his positions. 'Abd
al-'Aziz arranged that notaries should be present at his majlis,
making the condition that they should not be present at the
majlis of his cousin. In this manner the people were quite
uncertain what to do. [p. 254, 1] If one party brought a case
before al-Husain, his opponent would bring the case before 'Abd
al-'Aziz. When 'Abd al-'Aziz was present in the jami', the
place of al-Husain was quite deserted, so that the matter was
much talked about. Al-Hakim then wrote a diploma in his
own hand to the effect that trial cases should be brought only
before al-Husain. He ordered that no one should register.
judical decrees upon the authority of anyone else; and that if
anyone summoned a litigant who had already brought his case
before al-Husain, no one else could take chai-ge of it.1 This
diploma was read before the assembly.2 At this the cadi's
heart rejoiced. His growing prominence did not cease until he
reached the highest point of glory ; so that he compelled the
notaries to be present at his own house and in the jami', and
whenever one of them absented himself he had him punished.3
It was his custom to have the contents of the documents he
was to sign read in his presence before he added his own signa-
ture.

In spite of all this (harshness) he was very kind to men of
learning. He used to reward them with flour and barley, etc.,
and would send them garments and other things. This lasted
until al-Hakim ordered him to be removed from oflice in Rama-
dan 394. He knew nothing of this, remaining in his house,
until someone came and told him that his cousin 'Abd al-'Aziz
had been made cadi. He refused to believe it until it was sub-
stantiated. Thereupon he shut his door and remained in his
house. His fear waxed [p. 255, 1] until on Muharram 6, al-

1 For the meaning of Ije*.! .jX*-} Y see the instances cited by de
Goeje in the glossary to Tabari, p. CDLXXXIX.
s SLo, de Goeje, ibid., p. CDXC.

3 Ju^>. ? It is possible that the reading should be x
1X2*. »j' , i. e., he imposed a small fine.

286 R. J. H. Gottheil, [1906.

Hakim gave orders that he should be taken upon an ass in broad
daylight and imprisoned ; and at the beginning of the year 395
he was beheaded together with Abu Tahir al-Maghazill and the
muezzin of the kasr. The bodies of the three were then burned
near to the Bab al-Futuh. One of the things that led al-Hakim
to disavow him was the story of the man whom the chief of
police had beaten so that he died, as related previously.

Ibrahim ibn al-Rakik,1 in his history of North Africa, has
related the story of this al-Husain and al-Hakim. He says (in
the exact words): "And al-Hakim killed his cadi, Husain ibn
'All, and had him burned in fire." It is said that one of the
reasons of his killing him was that al-Hakim had been very
liberal to him;2 but had made the condition that he should keep
his hands off the people's money. A certain man who had a
grievance sent a paper to al-Hakim, in which he told him that
when his father died he had left him 20,000 dinars, and that it
had been placed in the diwan of the cadi Husain. He (the son)
was living for some time upon this money. He had come
one day and asked for some of it ; but the cadi told him that all
that his father had left was spent. Al-Hakim summoned the
cadi and showed him the complaint. The cadi answered just as
the complainant had averred, adding that whatever had been
left by the man's father had been spent on the man's living.
Al-Hakim at once ordered the books of the cadi dealing with
this matter to be brought. When this was done, al-Hakim
looked up the accounts of the man, and it turned out that he
had received only a little of the actual sum. Most of it was
found to be still due him. Al-Hakim enumerated to the cadi
the high offices he had given him, the various gifts and honors,
and his having made the condition that al-Husain should not
touch3 the people's money. He was afraid and terrified, and
said, "Forgive me, and I shall do better." He went away

1 Abu Ishak Ibrahim ibn al-Kasim al-Katib al-Kairuani al-Rakik al-
Nadim (Brockelmann, i, 155). His Ta'rlkh Kairudn is not mentioned
by Brockelmann ; but it is cited by Nuwairl, 'Idhari, Makrizi, Makkari,
Ibn Khaldun, Hajl Khalifa, etc. See Carl H. Becker, Beitrage zur
Gesch. Aegyptens, i, 9. The present citation shows that Becker is right
in placing his death later than 383 A.H. (the date given by Brockelmann).

2 Cf . the expression : XxJUs. &AA£. !^o .

3 Reading with Ms. 5893

Vol. xxvii.] A Distinguished Family of Fatimide Cadis. 287

with the man and paid him what was owing in the presence
of witnesses. But al-Hfikim bore him a grudge and had him
thrown into prison. Then he was taken out upon an ass in
broad daylight, the people looking on and following him until he
came to the loggia, where he was beheaded and his body burnt.

He held the office of cadi for five years, seven months and
eleven days. Al-Musabbihi says that he once pronounced the
li'an between a drunken man and his wife in the Jami' al-'Atik,
which was without precedent under the 'Ubaidites. 'Al-IIakim
gave to the aforementioned cadi as a fief a house in the vicinity of
the Khallj al-Hakiml. ' When the Nile was high he went (in a
boat) to this house ; the notaries came to him by land upon their
donkeys. Then he rode from it to the Kasr and returned ; after
which he. went to his dwelling-place in the Dar al-Hamra.

'Abd al-'Aziz ibn Muhammad ibn al-Nu'man ibn Muhammad
ibn al-JVJansur ibn Ahmad ibn Hayyun al-Kairuani was an Isma'I-
lian of the fourth century. He was born on the first of Rabl' 1,
355. He was appointed cadi on Thursday, Ramadan 16, 394,
and the revision of judgments was given in his power. He was
invested in the usual way, being carried upon a mule, while two
other mules were led before him, and a trunk filled with garments
was carried in front of him. He entered the jtirui', a large
assembly being present. His diploma was read out from the
pulpit. His first act as judge was to dismiss all the notaries
whom his uncle al-Husain had been accustomed to receive, with
the exception of Sharaf ibn Muhammad al-Makri, whom he
appointed to write down his decisions and his law cases. In
his protocols the following was his title:9 "The Chief cadi 'Abd

1 The Great Canal, variously called Khallj Misr, Khallj Amir al-Mii1-
minln, Khallj al-Hakiml, etc.; see al-Khifat ii, 138; Siyuti, Husn, i,
76 ; Ibn lyas, Td'rlkh Misr, p. 163, and de Sacy, Relation de CEgypte
par Abd-Allalif, p. 419, note 11.

1 The titles here given are of interest. He is called the " Cadi of 'Abd
Allah,'' with reference to the ShPite pretensions of descent from 'All

who is the jJLM 1J« (Van Berchem, /. c., p. 43). Mansur is part of the
caliph's name : Mansur abu 'All. The formula auJLfr asJJ! iiLJLo
«2Lj| J^ » to which is usually added

(ibid., p. 25 et al.), has a like reference. When al-Jauhar came to Fus-
t§t, he had the following words added at the end of the Khu^ba : " O
my God! bless Muhammad the chosen. All the accepted, Fatima the
pure, and al-Hasan and al-Husain, the two grandsons of the Apostle ;

288 7?. J. H. Gottheil, [1906.

al-'Aziz cadi of 'Abdallah and of his representative Mansfir-
Abfi 'All the Inifim al-Hakim, commander of the faithful — may
God bless him and his pure fathers — over Al-Kahira of al-Mu'izz,
Misr, Alexandria, the Two Holy Places, the districts of Syria,
al-Rahba, al-Rakka, the Maghrib, together with its provinces,

them whom thou hast freed from stain and thoroughly purified (^.5 jj|

&-U! ijJ*^)* ° mJ G°d! bless

the pure Imams, ancestors of the Commander of the believers (IlaJLJ!

^.XAX>._J! -x/o! AJ! yyJsLUJi SL^yi X.C. Jua.)." In the interesting
bit of Genizah poetry by one Solomon ben Joseph ha-Kohen (published
by Julius H. Greenstone in AJSL., January, 1906), the Fa^imids are
also called D^^llHDn (line 8). There are a number of such references
to peculiar Muhammadan titles and expressions, e. g., 1. 5 "l^Orf

<in line 9 D'OIN contains

a play upon the same word) ; 8b |

VI

1. 14a

cf. j»jL^JM oL^u, ; l. 20 ?pr = ^SAXO ; 113 Q'l

\^

sLojU! . I" a notarial document written for the Chief Cadi al-Kasim ibn
'Abd al-'Aziz ibn al-Nu'man (an account of which will be published in
the JQR. for April, 1907) and referring to the rebuilding of one of
the old synagogues.in Cairo, the full title of al-Mustansir is given. It is
interesting to compare the extent of the caliph's rule there mentioned
with that to be found in the diploma of his cadi. He is described as :

| . In the memoir mentioned

above, I have given all necessary explanations. Ibn Zulak in his

^£U6 va^lj' cjU5" (Paris Ms. 1817, p. 47a) says : i«J^«
/ ^ /

jjiA> —.' >*LM<J!

^5 I

/-^ ^;L°5

^ Vol. xxvii.] A Distinguished Family of Fatimide Cadis. 289

as well as whatsoever God has given into his power, and he has
made easy to be conquered by the Commander of the Faithful
(in the countries of the East and West)." '

And 'Abd al-'Aziz nominated to be his successor as judge
Malik ibn Sa'Id al-Fariki, and in [matters relating to] petitions
Ibn Abi 'Awwam. The notaries whom he had not received
hung around his door; so he sent to them [saying]: "Court
duties have increased greatly upon me and I shall need your
assistance in receiving testimony. Each one of you must
attend to his business; and whenever I need one of you for tes-
timony, I will make an appointment with him." Then they left
him. [p. 258, 1] But on the 17th day of Dhul-Ka'da he had them
come and made them take an oath that they had made no efforts
to find employment as notaries under his uncle, that they had
neither bribed him nor otherwise induced him to appoint them.8
They took the oath in regard to this, and he received them.

Al-Hakim caused 'Abd al-'Aziz to mount the minbar with
him upon Fridays and upon festive occasions, as had been the
custom of his predecessors. His power in legal matters extended
and his station became exalted. He took his seat in the jfimi'
and commenced to expound the work of his grandfather entitled
" The Basal Distinctions of the [Different] Schools of Law"
During his occupancy al-Hakim made over to him the care of
the Dar al-'Ilm3 which he had instituted. Al-Hakim [namely]

Al-Kuda'i, OvLsLjl (J*^ <->US" (Paris Ms. 1490,
fol. 141b) : xj.tXg+JI xJjJ ScXxij . ,j-o<X£jt ,jLo

UuLo

1 These words are added from Ms. 2152.

2 The Ms. reading aJ (or t^^c) IjtXcV^ »<X«£j Vj seems impossible.
I suggest the reading : Sjvil Vj ^j-1*1) ^5 » and nave translated accord-

ingly.

3 The Dar al-Ilm, or " House of Science," was founded in order to
propagate Shi'ite teaching in Egypt. Al-MakrizI (Khijat i. 458) has a
circumstantial account of its foundation drawn from al-Musabbihl,
who is evidently the authority followed by Ibn Hajar. The Dar al-
' llm was opened on the 10th of the second Jumada 395 A.H. and
was closed by Ibn 'Abd al-Tahir al-Af4al ibn Amir al-Juyush in the
sixth centuiy A.H. See, also, Ibn Khallikan, tr. de Slane, vol. i, p. xxix.

VOL. XXVII. 20

290 7?. J. H. Gottheil, [1906.

had built and arranged it, and had placed in it many scientific
works, throwing it open to the Fakihs, allowing them to sit in
it doing whatsoever they pleased — copying, studying or read-
ing; after it had been furnished, the hangings placed upon the
doors, and the necessary arrangements made for its attendants
and for the servants in charge of the furniture.

This 'Abd al-'Aziz was appointed to sit and to converse with
al-Hakim. The cadi found it necessary to give his older son al-
Kasim pei-mission to participate in his legal functions at the
jami', where he sat to hear cases and to decide disputes. Peo-
ple were accustomed to take cases from him to his father and
from his father to him. His younger son he ordered [p. 259, 1] to
verify1 the documents which people brought, and to render deci-
sions in a court held in his own dwelling. In addition, al-Hakim
appointed him ('Abd al-'Aziz) to administer the estate of his
cousin Husain ibn 'All ibn al-Nu'man, after that one had been
killed ; so he took charge of all his property. He did the same
with the estate of Abu Mansur al-Jauzi, one of the prominent
men of his reign. He preferred the cadi as prayer-leader over
a number of his assistants ; the custom having come into vogue
that the caliph alone should lead them in prayer. He com-
manded him to forbid men and women to promenade in the
streets on the 'Ashura festival ;2 it being their (i. e. Alid) cus-
tom to send out women and others who poured forth lamenta-
tions, who wept for al-Husain, and who chanted dirges in the
streets. The crowds were accustomed to stretch forth their
hands to the goods of the merchants. This having come to the
ear of al-Hakim, he commanded the cadi to forbid their going
about in the streets and [to order] that they should only lament
and chant in the open country.3

It happened that a certain Kutami 4 owed a certain sum but
refused to pay it. It having been in his possession during a bad

1 Or "to register."

2 On the 'Ashura festival, see the data which I have collected in the
Jewish Encyclopedia s. v.

3 1. e. the space between Cairo and Jabal al-Mukattam, called in al-
Makrlzi's time "The Smaller Karafa." See al-Khifat, ii, p. 455.

4 The Kutama were a Berber tribe. 'Ubaid Allah, the founder of the
Fatimid dynasty, was a Kutami. See al-Istakhrl, ed. de Goeje, p. 39,
below.

Vol. xxvii.] A Distinguished Family of F<iti,,<i<l< (',«?;*. 291

year,1 he had appropriated it for himself. The case came to the
attention of the cadi, who sent a messenger to the man ; but [the
messenger] was spurned. The matter was then brought to the
attention of al-Hakim, who ordered the Kutami to be dragged
before the cadi in Fustat. Then he was brought on foot to al-
Kahira and compelled to give up that which was due. Al-1 Ifikim
[also] entrusted to the cadi the care of the mosques, the hand-
ling of the wakfs, the collection of the products [belonging to
them] and the application of them [p. 260, 1] as he should think
well. This he did to the best of his ability, detailing for the
purpose two tellers who should pass, upon accounts.

The cadi betrothed his two sons to the daughters of the ka'id
Fadl ibn Salih. The ceremony was held in the castle, the mar-
riage contract being for four thousand dinars which al-Hakim
accorded from the public treasury. He bestowed upon them
ready-made robes of honor and sixteen double pieces of cloth.
The two were carried upon saddled mules, similar ones being
led before them.

The cadi was severe in his judgment ; his authority was great,
and he exercised authority over all the people of the kingdom.
He gave orders that those notaries who should absent themselves
in the morning up to the time of the majlis on Mondays and
Thursdays should be mulcted in a heavy fine. His delegate at
court, Malik ibn Sa'id, asked him to appoint al-Khalil ibn al-
Khalil in his place, as something had happened to him prevent-
ing him from riding or from going to court. This request
he granted; though this had not been permitted to any other,
that a delegate should name his own substitute in the city.

Al-Musabbihl relates in his history — when detailing the events
of the year 397 [A.H.] — the following circumstances, the gist
of which is: 'All ibn Sulaiman al-Munajjim,2 one of the inti-
mates of the commander-in-chief al-Hasan [p. 261, 1] ibn Jauhar.
told him that the cadi was visiting al-Husain ibn Jauhar the
kii'id in his dwelling upon one of the Christian's fast days.
He found there Abu al-Hasan al-RasI,3 al-Munajjira and their

1 The reading in the text is difficult. In lieu of it, I would suggest

Or " the astrologer."
3 The reading of the name is not clear in the Mss.; but see al-Dhahabi,
al-Tanblh, p. 246 ; Kosegarten, Chrestomathie, p. 121.

292 R. J. IL Gottheil, [1906.

attendants. A servant came in to say that Abu Ya'kub al-Kistas,
the physician, was at the door. He was invited to come in, the
company being at table. They made him welcome and a number
of dishes were placed before him. Then the table was cleared
and drinks were brought, also the fruits and scents belonging
thereto. They fell to until they became drunk. The cadi then
went away ; while the kfi'id and al-Rasi fell asleep. Abu Ya'kub,
the physician, remained in a portico which he had built in this
place — the portico overlooked a large stretch of water1 — drink-
ing and enjoying himself until he was overcome by drunken-
ness. They then went out looking for his mule. The mule of
al-Rasi was brought, but he refused to mount it. The servants
begged him to return to his place until his own mule should be
brought. So he went back to where al-Rasi was, and slept at
his side. Then one of the- lackeys came and lifted the curtain,
looking for the two. He saw al-Rasi but did not see Abu
Ya'kub; so he entered and searched for him; and [in the
end] caught sight of the tail of his garment in the water. He
called a lackey who knew how to swim ; this one jumped into
the water, and found Abu Ya'kub with his garments rolled
around his face and sunk in the water. The servants sent word
to the kfi'id, summoned the cadi, and waked up al-Rasi. [p. 262, 1]
The [whole] matter was most disagreeable to them, as they knew
the consideration al-Hfikim had for al-Ya'kub. They begged me
to inform al-Hakim of what had happened. I went to him and
told him that Abu Ya'kub had gotten up daring the night while
in a stupor, and had jumped into the river; and that when the
lackey got to him he found him caught in his clothing and
drowned. This grieved al-Hakim, who appeared to be very
much affected. He had the matter looked into. The exact cir-
cumstances Avere related to him, but he shook his head.

The cadi, al-Rasi and the ka'id, however, had come to the
castle on foot with fine turbans on their heads. Al-Hakim sum-
moned them [to his presence]. They swore and affirmed upon
oath that they had had nothing to do with the whole affair. The
ka'id and the cadi called al-Rasi to witness, and he testified to
their innocence. Oi'ders were then given that the body should

1 The description confirms Dozy's suggestion (ii. 41) that the al*vLb
was a " portique ouvert."

Vol. xxvii.] A Distinguished Family of Fatimide Cadis. 293

be prepared and buried. This occurred towards the end of the
year 397 [A.H.].

On Thursday, the middle of Rajab 398, the report was spread
abroad that 'Abd al-'AzIz, the cadi, had been deposed and that
his successor, Malik ibn Sa'id, had been confirmed. When the
morning came, he did not go to the court until near mid-day.
Then he did go ; held court, led the mid-day prayers and went
off all alone, without chamberlain or footman, until he came to
his dwelling place. As the day commenced to decline, certain
people went around to all the .chief men telling them to collect
in the castle on the morrow, [p. 263, 1] So they all assembled
before Malik ibn Sa'id, who was invested with the various func-
tions with which 'Abd al-'Aziz had .been invested — whose term
of office had [thus] lasted three years, nine months and twenty-
eight days.

Al-Musabbilri says: 'Abd al-'Aziz dismissed from office
thirteen persons while he sat as appeal judge, and two during
his cadiship. After his deposition, 'Abd al-'Aziz was in the
habit of going to the castle alone, though constantly expecting
to be killed. On the 13th of the second Jumada, 399 [A.H.J,
the ka'id Husain ibn Jauhar and the cadi went out riding
according to their custom, and they came back without any
incident having occurred.1 Then [al-Hakim] sent for them;
and 'Abd al-'Aziz going first, was clapped into prison. His
servant returned home [alone] with his mule; whereupon the
kfi'id and his son hid themselves. But the door of their house
was broken in and al-Hfikim gave stringent orders that they
should be brought. This being found impossible, al-Hakim
gave word that 'Abd al-'Aziz should be released, who returned
to his dwelling place. The people were already making prepa-
rations for his funeral, but he quieted them. The shop-keepers
had also closed their booths; these he ordered to be opened
again. After three days, the ka'id went to al-Hakim unmo-
lested. Splendid robes were presented to him and to 'Abd al-
'Aziz; before them many garments being carried. They were
led [in state] upon two horses, many horsemen preceding them.
Then al-Hakim gave back to 'Abd al-'Aziz the hearing of appeal
cases. His certificate was read out; he was clothed with

1 UJLwwJ ; or " having greeted " [al-Hakim.]

294 R. J. H. Gottheil, [1906.

ready-made garments [p. 264, 1] and with thetailasan. He was
led riding upon a mule ; before him was another and in front of
him was carried a chest full of garments. He was [thus] con-
firmed1 on Safar 9th, 400 [A.H.] The revenues of a fief were
given to him, and upon the door of his dwelling-place a plaque
was put with the name of the diwan. On the last days of
Ramadan the cadi's sons married the daughters of the ka'id to
whom they had been betrothed.

In the latter part of Muharram 401 [A.H.] the cadi and the
ka'id became suspicious of perfidy on the part of al-Hakim
towards them; and on Safar the 9th the cadi, the commander-
in-chief Husain, their followers and their friends took to flight,
carrying with them much .of their wealth, and went in the
direction of Dujwa.4 When al-Hakim heard of this, he put
seals on their dwellings and ordered Malik ibn Sa'id al-Furik!
to ride to the dwellings of the cadi and of al-Husain, to seize
whatever he might find there and to carry it off. The cadi and
the ka'id kept in hiding until Muharrain the 6th, 401 [A.H.]
when they appeared, a safe-conduct having been written for
them. They remained at their posts until Friday, the 12th of
Jumada, on which day they were present at their posts and then
returned. At once al-Hakim sent to them and they came back ;
whereupon a band of Turks killed both in the vestibule. The
seal was .at once put [p. 265, 1] on their dwellings, their houses
were immediately surrounded, and they passed away unavenged.
Many of their followers were seized and fined.

'Abd al-'Aziz was learned in the canon-law of the ImumI
rite, as his whole family had been, especially his grandfather.
The Sheikh 'Imad al-Dln ibn al-Kathir3 attributed to him the
authorship of a work entitled Al-Balagh al-Akbar wal-JVamus
al-Aizam dealing with the principles of religion. But in this
he was mistaken ; for it was a work composed by his father giv-

1 Reading JLX.*J^ .

2 Dujwa or Dijwa, six parasangs from Fus^at, Yfikut, ii, 555, in the
Sharkiyya province. This can hardly be the present ^£»— > 3 » marked
in the Egyptian Postal Guide (Maslahat al-Busta, Cairo 1906, p. 268) as
being in the Kalyubiyya province.

3 Born 1301, died 1373. The reference may be to his large historical
work, Al-Biddya wal-Nihdya, Brockelmann, ii, 49.

Vol. xxvii.] A Distinguished Family of Fatimide Cadis. 295

ing the ideas of al-Nu'man, that one's father. Ibn Kathlr says
that the cadi Abu Bakr al-Bakilam1 wrote a refutation of this
work. Ibn Kathir adds "It contains heterodox ideas, the like
of which Iblis himself would not have conceived." Such were
his words.

Kasim ibn Abd al-'Aziz ibn Muhammad ibn al-Nu'man al-Ma-
ghribl belonging to the Imaml sect, of the fifth century. He
was called Abu Muhammad, and was put in office after Ibn
Abi al-'Awwam" on Sunday, the 4th of the first Jumada, in the
year 418. 3 His diploma was published in the kasr and in the
jami' of Fustat. He received the title Kadi al-Kudat, Thikat al-
Daula, Amin al-A'imma, Sharaf al-Ahkam, Jalal al-Islam.4 He
exercised his functions until he was deposed on Sunday the 25th
of Rajab in the year 419, he having lasted- for one year, two
months and some days. This was his first tenure of office. In
his place was appointed 'Abd al-Hakim ibn Sa'id ibn Malik
al-Fariki. On the 6th of Dhul-ka'da, in the year 427, Kasim
was re-appointed to office, presided over civil [p. 266, 1] and
criminal cases, and (also) over the preaching. In this period of
office al-Kuda'i was appointed and became his locum tenens in
this, his second period. Kasim was not praised for his con-
duct, although his term of office lasted for a long time, until he

1 Abu Bakr Muhammad ibn al-Tayyib ibn Muhammad ibn Ja'far ibn
al-Kasim al-Bakilam, died 1013. See Ibn Khallikan, tr. de Slane, ii.
671. Perhaps the work referred to is his Kashf Asrar al-Batinlyya
mentioned by Haji Khalifa, v, 199.

2 I. e., Abu al-' Abbas Ahmad ibn Muhammad ibn 'Abd Allah ibn Abi
'Awwam.

3 Al-Musabbihi, in treating of the year 415, mentions the fact that al-
Tahir had al-Kasim ascend the minbar with him. He is here called only
"Chief Preacher." Becker, Beitrdge, i, 72, 4.

4 The synagogue document above referred to (of the year 429 A.H.)

gives his title as follows : ..v-yot sLcjJI c^^

!. Al-SiyutI (vol. ii, p. 102)
says of our cadi : xJ. jj! XAJ. sLfr jJ! -^ sLoJlM

296 Gottheil, A Distinguished Family, etc. [1906.

was deposed in al-Muha.rr'am 441 ; this second terra having
endured thirteen years, one month and four days.1 The verses
in which both he and Ibn 'Abd al-Hakim al-Fariki were ridi-
culed have already been cited.2

Abu al-Kasim ibn Muhammad ibn al-Nu'mah is the same as
Ibn 'Abd al-'Aziz previously mentioned.

AN ADDITIONAL NOTE AND SOME CORRECTIONS.

P. 224. I am not quite correct in saying that the history of the Egyp-
tian cadis was first written by al-Kindi. Professor Torrey calls my
attention to the fact that this was done before him by Abu al-Kasim
'Abd al-Rahman ibn 'Abd al-Hakam (died 257 A.H.). This was not
apparent from any of the reference books; but see now Torrey, "The
Muhammedan Conquest of Egypt and North Africa," in Biblical and
Semitic Studies (Yale Bicentennial Publications), N. Y. 1901, p. 279 :
" Appended to the history proper is a collection of brief biographies of
the qddls of Egypt, from the Conquest down nearly to the author's own
time.'' It forms the sixth of the seven parts of the Futuh Misr. As it
stops at the year 246 A.H., it is evident that al-Kindi has based his
treatise upon these notes of Ibn 'Abd al-Hakam. I purpose to edit
al-Kindfs account of the cadis.

P. 229, note 4 read Ravaisse.

P. 245, 9 read I g-^v ,

P. 239, 13 read xb^xiJl .
P. 240, 1 read LI .

P. 248, 9 read KUi&> .
P. 248, 13 read (jWvJ .

P. 242, 1 read vJsUxi .
P. 242, 4 read ojC*,. .

P. 250, 8 read xcJL^XA-vJ. .
J

P. 251, 4 read .v^JLuJ. .

P. 243, 11 read gLdftJ! .

P. 254, 5, 10 read J^J .

P. 245, 2 read ^.oLs .

(S

P. 245, 3 read xiLiij .

P. 254, 10 read M^ .
P. 256, 16 read JoxJ! .

1 Al-Makrizi (al-Khltat, vol. i, p. 355) says that he held office for four-
teen years ; but this is probably not meant to be absolutely exact.
* In some other biography of this same work.

Supplement to the Old- Babylonian Vocabularies. — By R. J.
LAU, Ph.D., Columbia University, New York City.

IN 1896 Columbia University in New York City came into
possession of 455 Babylonian clay-tablets, 258 of which are a
part of the so-called Telloh tablets, discovered by Ernst de
Sarzec in 1894-L895. They can be assigned to the fourth
dynasty of Ur (2750-2550 B. C.), both because of the given
dates and for paleographic reasons; though a number of the
tablets contain only the day and month, else no date at all.

While copying, transcribing and translating these inscriptions
I met with some expressions which, I believe, have not yet been
translated. And when I later compared the 'Sign-List and
Glossary' of my book, Ancient Babylonian Temple Records
(soon to appear), with George A. Reissners Tempelurkunden,
I found that the Columbia University Collection contained
words and phrases which have been left untranslated by other
authors, or which do not occur on any tablets so far published.

I herewith offer the following list as a supplement to the
existing vocabularies :

AD.E

BA

r s

298

M. J. Lau,

[1906.

"DUK

-+GrUN(J

- /•

s (p ~?)J -f-

T)u MU

KU.MAL,
/ru/LWi Sl&M-

DA.
G'EMS HXR.HfiR

(A IS) vrv'/n/ntc/i' ~urdo4/(U& fa U

SAJauj} i /u. t .;

HU KAK
KA

Vol. xxvii.] Supplement to the Old- Babylonian Vocabularies. 299

/ou JoJ(/t/UtiL<t> dodL/Vsmt

a A B ^/naUMJ') /nrirfJy AM Jia/nds(<S(/ ^

f<A

&

A/A QlS.Nt
A/A »f/^

/cus/tnaM; fra/ijuAts Jf&t&rttoj *j.i,t

A/, l., &st '/d /I'tT^fo'txrujwi .

MU . . .

) =

V • / <>

,w<^ (M u = 3

300

72. J. Lau, Supplement, etc.

[1906.

$1 Nl. IB NIN S/D.Atir

Abel totf) in the Bible.— By E. J. LAU, Ph.D., Columbia
University, New York City.

Siegfried and Stade (Hebraisches Handwdrterbuch, p. 5a)
translate this word 'grassy plain, pasturage (Aue, Trift) ;'
others have suggested that it should be read 73K •

It neither means 'grassy plain,' nor must it be changed to
O^ f°r tne following reasons:

1. On such an ^DK the Israelites had placed the ' Ark of Jahve '

(1 Sam. 6, 18: "fT p^V^ H^ in^H 10J n^ll-l ^N)-
If this ^5^ na(^ been a 'grassy plain,' the text would read
m 1IT3n "KPN. instead of JTty IIT^n "K5>N : it must

T • • v - : T v *r • • v - :

therefore have been an object higher than the ground itself.
According to verses 14 and 15 it was an H^IIJJ ON- ' a
large stone,' which still stood in the field of Jehoshua
(v. 18, last clause) in the time of the writer of the book of
Samuel.

2. Other passages in which /Dl^ occurs seem to prove that these

stones were placed in certain localities

a) to commemorate well-known events of the past:

a) the DHVP^N. 'the ^K of the Egyptians,'
where the Israelites (called here Egyptians)
mourned for Joseph; cf. Gen. 50, 11.

B) the n^inD^N. 'the *?2N of the dance,' which

T : •• T •• T

had been placed in memory of a certain 'great
rejoicing ' of the people ; Judges 7, 22 ; 1 Kings
14, 12; 19, 16.
y) the rO^E n>3 "^K > ' the ^DN at Beth Maa-

T ^ : - •• T •• T

chah;' 1 Kings 15, 20; 2 Kings 15, 29; 2 Sam.
20, 14, 15.

b) to mark possession ; with a signification similar to that
of the Assyrian kudurru, 'boundary-stone.'

a) D.*0 *??£ • ' the ^N of (at) the water(s) ;' 2
Chron. 16, 14.

302 R. >T. Lau, Abel (tf) in the Jtible. [1906.

/8) DWnN. 'the Dtf of (at) the locust-trees;'

Nu. 33, 49.
y) D'0"D ^SN • ' the ^DN of (at) the vine-yards ;'

• T : " T •• T

Judg. 11, 33.

This ^5^ stone'" was not merely a boundary stone, but one
that marked 'possession.' The fact that the word occurs only
in the singular goes far to prove, that

1. only one stone was placed on the land, at the waters, in the

grove of locust-trees, or in the vineyards, mentioned above ;

2. that most likely it was larger than a common boundary-stone,

but lower than the cart on which the ' ark ' was moved.

3. A further proof for this assertion can be adduced from

the Assyrian ablu, iblu:

a) Nebuchadnezzar styles himself (VR. 55, 5): misir
kudurreti, muMnu able, ' protector of the boundary-
stones, and establisher of the able (stones).' Accord-
ing to this passage the kudurru was different from
the ablu.

b) Nabopalassar says (OBI. I, col. II, 28-31 : amelDI3f.

GAL. E istattum (=istenis) ibU ukinnu1 'the mas-

ter-builders determined the ibleS Here ible must

mean not merely the boundaries, but rather the e.>'ti nt

of the boundaries in either direction, that is they

determined where the ible-stones should be placed,

which marked the extent of the boundaries.

According to these two passages the Assyrians made use of

more than one ablu or iblu, which were not the same as the

kudurre, for the words occur only in the plui-al. The Hebrew

^DK °ccurs only in the singular, and was placed 1) to commem-

orate a certain event; or 2) to signify possession.

The Pi'lel in Hebrew. — By Louis B. WOLFENSON, Johns Hop-
kins University, Baltimore, Md.

• ' /

IN the Semitic languages the great majority of words are
derived from triconsonantal roots. There are a number of
pluriconsonantal roots,1 e. g., DD"O (impf. Qal with suffix Ps.
80, 14), 'eat off,' .jrnflV 'frog'; Arab. ^x+Z qam'-al 'sprout;

be chief; Syr. ^nji 'hasten'; Eth. DIM lanbasa (J^-A-A_£-
lanbas from (j***^) ' lion,' etc. ; but these are in nearly all cases
derived from triconsonantal roots in various ways.2 According
to some authorities all triconsonantal roots are in their turn
derived from biconsonantal roots.3 The biconsonantal roots,
however, they regard as altogether prehistoric, and all words in
the historic stages of the languages are looked upon as derived
from triconsonantal stems. The shorter biconsonantal form of
the verbs *V'J^4 and $"$ are thus considered to be the result
of elision and contraction of triconsonantal ones.

1 The number of pluriconsonantal roots in Hebrew is small ; in Syriac
the number is larger, while in Arabic and especially in Ethiopic they are
relatively numerous.

2 For the ways in which these formations are developed cf. Gesenius-
Kautzsch27, §30, p. q.; Dillmann, Aeth. Gram.*. Leipzig, 1899, §§71-73,
77, 78 ; Noldeke, Syr. Gram.2, 'Leipzig, 1898, § 180 ; and the special treat-
ises of F. G. Schwartzlose, De Linguae Arabicae Verborum Plurilittero-
rum Derivatione, Berolini, 1854 ; Stade, Ueber den Ursprung der Mehr-
lautigen Thatworter der Ge'ezsprache, Leipzig, 1871 ; Martin Hartmann,
Die Pluriliteralbildung in Semitischen Sprachen, Halle, 1875 [only the
Erster Theil ; Bildungen durch wiederholung des letzten Radicales am
Schluss und des ersten nach dem zweiten has appeared] ; Siegmund
Fraenkel, Beitrdge zur Erklarung der Mehrlautigen Bildungen im Ara-
bischen. Leiden, 1878.

3Cf. Ed. Konig, Lehrgebdude der Hebr. Sprache, Leipzig, 1881, 1895,
II1, § 119, "3 b), c) (p. 370 ff.); and contrast Gesenius-Kautzsch27, p. 99, n. 1.
It is most likely that originally all roots were not biconsonantal, but that
there were also triconsonantal ones; cf. Delitzsch, Studien uber Indo-
germanisch-Semitische Wurzelverwandtschaft, Leipzig, 1873, p. 70.

4 Verbs *V tt is used as a convenient symbol meaning verbs mediae u,
mediae i, following Konig, who uses also *Y'fl similarly.

304 L. B. Wolfenson, [1906.

This view of these verbs is that of the national Arabic gram-
marians, and it is supported by the many secondarily regular
forms in their language. In Hebrew, however, which is in some
respects more primitive than Arabic,1 the conditions are differ-
ent, and the earliest Hebrew grammarians and lexicographers of
tlie Middle Ages did not hold this. view. They believed that in
Hebrew there are biconsouantal2 and even uniconsonantal roots
in the case of certain weak roots like PDJ3, etc. The explana-
tion that the shorter forms of the verbs *V'J? and J?"J7 are con-
tractions,, is based on the Arabic view, and was first introduced
in Hebrew by the grammarian and lexicographer Hayyuj^ liv-
ing at Cordova, Spain, in the latter half of the 10th century and
early part of the llth, who spoke and wrote Arabic, and applied
to Hebrew the principles of the Arabic language and the meth-
ods of the Arabic grammarians. His view of these verbs pre-
vailed until the last century, and is held even at the present
time by such a prominent grammarian as Ed. Konig,b as well as
by others of less note.

1 Cf. Gesenius-Kautzsch5", § 1, n.

2 This is shown by the arrangement of their lexicons. Thus, e. g., in
the lexicon of Menahem ben Saruk we find the root {^j^J treated under
the biconsonantal heading J^JJJ ; nit^ aQd Hfl^ both under

' an<* T^lDJ UQder V£ ; showing that the ultimate root of many so-
called weak roots was considered biconsonantal. See the edition of
Menahem's Lexicon by Herschell Filipowsky, Antiquissinmm Linguae
Hebraicae et Chaldaicae Lexicon . . . A Menahem ben Saruk . . .
Londini et Edinburgt MDCCCLIV, pp. [168]a, [I71]b, [144]% and cf. J.
Fiirst, Zur Geschichte der Hebrdischen Lexicographic, the Introduction
to his Hebrdisches u. Chald. Handworterbuch, Leipzig, 1863 (2d ed.),
p. xx.

3 Cf. the Lexicon of Menahem, p. [103]b, under "j for ,""O3 ; cf. also p.

[123]a, and [127]b for other examples of uniconsonantal roots (7 root of

nj» . D root of HDJ)-

4 His views on this subject are expounded in the two treatises called
the Kitdb al-Af-dl dawdt Huruf al-Lin, and the Kitab al-Af-al dawat
al-Mithlain; see the edition by Morris Jastrow, Jr., published under the
title "The Weak and Geminate Verbs in Hebrew, by .... Hayyuj,"
Leide, 1897, Preface, p. xi, and cf. Fiirst, 1. c., p. xxiv.

5 See his Lehrgebdude, I, §34 (p. 320 ff.), where the verbs V"V are
treated under the heading of Contracted Verbs. Cf . also Vorrede VII,
and pp. 479-81 with pp. 451-53. Of course the question of the ultimate
origin of these shorter verbs is not affected by this opinion. Thus both

Vol. xxvii.] The PMel in Hebrew. 305

However, beginning with J. Fiirst1 and Ewald," there has been
a constantly growing number of scholars who have regarded the
verbs *V'^ and $"$ not as contractions of triconsonantal forms,
but as developments of biconsonantal roots which were not
expanded to the triconsonantal forms as in other cases. Thus
Noldeke,3 Bottcher,4 A. Miiller,6 Stade,6 deLagarde,7 Friederich
Delitzsch,8 Zimmern,9 Kautzsch,10 Wellhausen,11 and others have

Konig (cf. p. 303, n. 3), and Mayer Lambert, who believes that the verbs
*V17 and y ' V are contractions of triconsonantal forms (cf. his article
" La Trilitteralite des Racines V"V et *\" \} ," in Revue des Etudes Juives
[REJJ, Tome xxxv, 1897, p. 203 ff.)> consider that these verbs (*y'^ and
yy as well as all other triconsonantal verbs are derived from original
biconsonantal roots. Cf. Mayer Lambert's paper in Semitic Studies
in Honor of Alex. Kohut, Berlin, 1897, p. 354-62, but contrast Gesenius-
Kautzsch27, p. 99, n. 1.

1 Cf. Lehrgebaude der aramdischen Idiome, Leipzig, 1835, §§ 91 (p. 81),
153 (p. 158). As far as I can find, no credit has been given Fiirst for
postulating the theory that the verbs *V'17 an^ 1^"17 are biconsonan-
tals, as his name is omitted in everything on this subject wjiich I have
seen, Ewald and Bottcher being the first scholars mentioned as holding
this view.

2 Cf. Lehrbuch der Hebraischen Sprache9, Gottingen, 1870, §§ 112, 113.

3 In a review of Olshausen's grammar in Benfey's Orient u. Occident,
I, 1862, p. 760 ff.; cf. Mandaische Gram., Halle, 1875, §87, and Beitrdge
zu semit. Sprachwissenschaft [BzsS.], Strassburg, 1904, p. 46.

4 Lehrbuch, 1866-68, §§1116f.; 1127 f.
6 In ZDMG. 33, 1879, pp. 698-700.

6 Hebr. Gram., 1879, pp. 109 ff., 138 ff.

7 Cf. Orientalia, II., Gottingen, 1880, p. 6 ; Ubersicht, Gottingen, 1889,
pp. 26, 27.

8 Assyrian Gram., Berlin, 1889, § 61, 1) ; § 115.

9 Vergl. Gram. d. semit. Sprachen, Berlin, 1898, § 50 b, c. ; § 51 b, c.

10 Gesenius-Kautzsch, Hebr. Gram.*\ §§ 67, 72.

11 Skizzen u. Vorarbeiten, VI, Berlin, 1899, p. 250-255. This article
Wellhausen says he wrote to explain especially the impfs. of the so-
called verbs vy, e. g., Q*jj^ , 7*3*. De Lagarde, however, had pre-

viously stated that these verbs were biconsonantal just as the verbs
called V'tt. In 1880 in his Orientalia, II., p. 6, he says : "die wurzeln
•JT7 und i]f halte ich garnicht fur dreiconsonantig, sondern — seit jaren
habe ich dies offentlich gelehrt — fur zweiconsonantig," and in his Uber-
sicht, p. 26, 27: "Ich glaube, dass es zweikonsonantige Wurzeln mit
ursprunglich langem Vokale gibt : j*Ls mit (•«JL> [• • • •]• Ich fiige jetzt

s °
hinzu, dass ich ^j mit .J^ parallel setze, etc." Apparently no notice

has been taken of these statements. Wellhausen does not refer to
VOL. xxvii. 21

306 L. B. Wolfenson, [1906.

considered these verbs as biconsonantal. It is A. Miiller, Stade,
and Wellhausen especially, who have most consistently carried
out the biconsonantal explanation.

According to their explanation, these verbs are derived from
original biconsonantal roots with a short characteristic vowel
between the two radicals, corresponding to the characteristic
vowel between the 2d and 3d radicals of triconsonantal verbs.
Under the influence of the prevailing triconsonantal types, the
biconsonantal forms of the verbs 'V'J? and ^"j^ were usually
amplified: the former, by lengthening the short characteristic
vowel between the two radicals,1 e. g., in Dip* the $ is from

Lagarde, nor does Noldeke in the reprint of his paper Die Verba ">]) im
Hebrdischen in BzsS., p. 34 ff., although Lagarde refers (Uebers., p. 26
below) to Noldeke's article (first published ZDMG. 37, 1883, p. 525 ff.),
and Noldeke accepts the explanation of Wellhausen (BzsS., p. 46).
Lagarde's explanation, however, is based on the assumption that the
verbs ^V'J/ had an originally long vowel between the two radicals (cf.
above). This same view is held by Ewald, Delitzsch, and Zimmern (cf.
II. cc.). Wellhausen, on the contrary, correctly explains these verbs as
derived from biconsonantal roots with an originally short characteristic
vowel, so that the i in Q*{£?* is lengthened (under the influence of the
longer, predominant triconsonantal forms) from I, just as the u in Q^*
is from u, and the 6 for d in J^i^* from #•

1 This lengthening takes place usually in forms in which the charac-
teristic vowel stood originally in an open syllable, e. g., Arabic +\ _ '•, ,

i' , IfcXili', etc., from original qdma, qdmdt, qdmu ; Syriac

^ Q.<aQc ? etc. But in Hebrew this lengthening did not take place in
the corresponding forms of the Qal perf., Qp , HOP (DD> £^3). etc->

IT T IT

being for qdm, qdmd (met, bos), with tone-long vowels, and so really=
qdm(d), qdmd (mlt, bus), etc., with heightening (not lengthening) in the
tone. The forms of the Qal act. part, are the same as those of the 3d
masc. sing, perf., hence qdm, met, bos, although the vowels are
unchangeable. Cf. F. R. Blake in JAOS. vol. xxii, 1901, p. 51, n. 3 ;
Wellhausen, Skizzen u. Vorarb. VI., p. 252; and contrast Gesenius-
Kautzsch", §72g.

In both Hebrew and Arabic an originally short characteristic vowel was
retained without lengthening when it occurred in a closed syllable.
Thus in the jussive and apocopated forms the original short vowel was not
lengthened because in a form expressing a command or the like it was

desirable to have as short a form as possible, e. g. , Qr})=idqom with

I T

later tone-long 6 from jo</ftm,Qp»^= iianaqom in which the original u

Vol. xxvii.] The PMel in Hebrew. 307

an original $, in /*J|* the i is from an original \\ and in NlD* the

6 is for d from $; the latter by doubling (not repeating) the

<
second radical j e. g., HDD from an original scibcit, etc., the

doubling being secondarily omitted in Hebrew in forms in which
the final vowels are dropped, e. g., ^D for sabb, cf. Arabic

"ifarra, etc.

appears as 6, ^y>=jfigel with tone-long e from iagil, etc. ; Arabic

" T

iaqul, _**.> iasir, Jyj iaz&l, etc. Also in forms having an afformative

beginning with a consonant, the characteristic vowel, occurring in a
closed syllable with a second consonant immediately following the
final radical, was not lengthened, no doubt on account of the firmness
resulting from the juxtaposition of two consonants without intervening

vowel, e. g., Arabic <&*+£ qumta, OA.AJ blnta ; Hebrew fipP » f1&'^}
(with 6 heightened from u in the tone): impf. ..vJLftJ iaqulna, .J^A^J

iaslrna, ..vyj iaz&lna; Hebrew nj^r) with o heightened (not

lengthened) from u (if it were lengthened it would become ft as in Qlf^

, I T

from iaqum), HJ 7Jfl witn ^ from *> and n^JOn with 6 (Qot 6 as is

T : •• T T T

stated, Ges.-Kautzsch27, § 72 k) obscured from a which was lengthened
from an original a on account of the quiescing of the }$ in an original
taba'na. [In the rarer n^iOfl (cf- Ges.-Kautzsch-1, § 76 g) the _i_ is

T V :

also 6 for a ; but here the d — as well as the ft in nJPOIDfl an(^ *ne * *n

TV) :

n^D^pn~arose through the lengthening of an original short vowel,

T -.•(•:

a (u, t), under the influence of the prevailing triconsonantal type, since
the root syllable is no longer closed when * ~^r intervenes before the affor-
mative J~fJ — .]

In such forms as these, in which the characteristic vowel occurs in a
closed syllable, some (e. g. , A. Miiller, ZDMG. 33, p. 699) are inclined to
think that this vowel was first lengthened and then shortened again in
a closed syllable, so that fttt), e. g., is shortened from *qdmta, which

T : |-

arose from qamta. This is apparently supported by Syriac *Sa." , etc. ,
Ethiopic ^^°h qomka, etc., with long vowel in a closed syllable. In
Ethiopic, however, the long vowel in the closed syllable is contrary to
rule. Cf. Praetorius, Gram. Aethiopica, 1886, § 15. The long vowel
here must be explained. In both Ethiopic and Syriac the long vowel is
best explained as due to the analogy of other forms in which the long
vowel occurs regularly in an open syllable, e. g., ASn" aSi-° etc.;
$OD qoma, fyao^ qomat, $OD* qomu, etc. The Hebrew forms like
npp are then to be explained as preserving the originally short vowel

unchanged, and are therefore more original.

308 L. B. Wolfenson, [1906.

This biconsonantal theory is the most natural explanation of
these classes of roots, and is the one most in accord with philo-
logical principles. For if the prevailing U'iconsonantal type of
root is in considerable part a development from a biconsonantal
state,1 it is more than likely that remains of this former state
should be preserved in the stages with which we are familar.
In language a new order of things is a growth, the older exist-
ing at least for a time beside the new, and it is not introduced
by unanimous agreement, as it were, of those using it. In all
languages in which a growth can be observed a certain number
of older forms are preserved. These older forms appear irregu-
lar in comparison with the prevailing types. To consider the

This conclusion is supported by the corresponding Arabic forms
qumta, o*-*J bmta, etc., in which the vowels are also short. The short-
ness is original. Their quality, however is secondary. One would
expect to find a in the root syllable, as in Hebrew. Wellhausen has cor-
rectly explained the u and I as due to the characteristic vowels ti and i
in the impf . ,*«JiJ iaqttm, -.^AXJ iatnn. Of course in the case of intrans.

verbs like JLb ' be long ' (oJlb tulta), Jk ' cease ' (oJv ziltci). it should

occasion no surprise that the original intrans. characteristic vowels are
retained. Noldeke, however, has questioned (BzsS., p. 46, n. 2) in this
connection : Why, if oJ\ zilta is the intrans. form, do we not find Jo\

zila instead of Jk zdla? By way of answer it will be recalled that such
intrans. forms are actually found dialectically; cf. Wright-deGoeje,
Arabic Gram.3 1, 1896, p. 83 D. In general, however, this form (Ju; )

became the passive in the case of trans, verbs (cf . F. R. Blake's paper.
The Internal Passive in Semitic, JAOS., vol. xxii, p. 51 ff.), and when
this took place the act. form jLi' prevailed also in the case of intrans.
verbs in forms in which the characteristic vowel occurs in an open syl-
lable (owx>lj> qdmat, \^a\Ji qdmti, etc,). In Hebrew also the trans, type
Dp DDP prevailed in many verbs that must have been originally

IT : |-

intrans. Only ]•}£ , JJJ'i^ , "Tjtf , and ^JJJ^ occur as intrans. forms in the
perf. The trans, form prevailed to such an extent that we find the
trans, vocalization in the case of forms of J^Q having an afformative

beginning with a consonant, e. g., HDD and not

All the forms of the verbs *y M may thus be satisfactorily explained
on the biconsonantal theory.
1 Cf. n. 3, p. 303.

Vol. xxvii.] The PMel in Hebrew. 309

shorter forms of the verbs *V'l? and yyi as contractions of
regular ones is unphilological2 — they are original biconsonantal
forms preserved in the historical stages of the Semitic lan-
guages.3

Accordingly, forms of these verbs with three distinct radicals
are a relatively later development. In fact, in case of the roots
^"y in Hebrew, verbal forms with consonantal4 1 and ' are very

1 I retain the symbols *\'y and V" V as being customary and conven-
ient, although they are, of course, inaccurate since there was properly
no radical in these verbs corresponding to y in ^yQ-

• In the case of the verbs *Y'J7 it is even impossible. For in the verbs
that actually have ^ as 2d radical we find the *) preserved as a radical
consonant which does not suffer contraction, and that too in just those
situations in which the advocates of the triconsonantal explanation of
the verbs *Y'tt say that * and ^ were elided or underwent contraction
e. g., m") I Sam. l6. 23, beside forms like QH in verbs Y'tt, }|T1|T

-T IT T| : •

beside forms like )iy\fp , etc., etc. Of verbs with *) as 3d radical there
occur (not including verbs tfn"1?) in Hebrew Jfljj, "Hlf, ^y , t\*\y >
"liy • fill? • IDS • Pm . and y\W - cf • below, n. 4. Accordingly there
is no reason why a contraction should have taken place in verbs \'y if
^ had been present originally as 3d radical any more than in the above
verbs. We must therefore conclude that no ^ was present. These verbs
(i. e., the so-called verbs )\'y ) were originally biconsonantal.

3 As Noldeke pointed out as early as 1862, saying (Orient u. Occid. I,
p. 760) : " Wir betrachten eben Wurzeln wie q&m, sab als werth voile
Uberreste einer Zeit in welche die Dreikonsonatigkeit noch nicht bestand. "
Accordingly these roots have only two radicals. In his more recent
statement (BzsS., p. 46 below, 47), however, viz., "Alle historischen
semitischen Sprachen behandeln hier doch die Vokalbuchstaben ^ und »
als Radicale," there is not a little inconsistency. According to this
remark, there are practically three radicals. This is open to grave
objections in fact, as Noldeke himself must needs grant, since he admits
that the Arabic forms cited by him in support of his statement may be
considered secondary.

4 Of course, forms of triconsonantal roots with ^ as second radical are
formed regularly, and in these ") appears as a consonant throughout,
e. g., ^W , rfr\y ; HIV • IITO* Is- 42' n ; etc., cf. n. 2. The gutturals

• - : T :- |T : •

in these roots have nothing whatever to do with the retention of the ^ as
a consonant, as Konig (Lehrgeb. I, p. 458) followed by Mayer Lambert
(REJ., xxxv, 1897, p. 211) supposes, since we find a large number of
roots ]"y which also have gutturals as radicals, and yet no consonantal
*) appears, as one would expect according to Konig, if it had once been

310 L. B. Wolfenson, [1906.

few, being limited to Pixels occurring only in the latest litera-
ture, so that they are really Aramaic forms and not Hebrew,
e- 'g-» "Tltf Ps- Up, 61; D*p Est. 9, 21, 29, 31; ID'p Esth. 9,
27, 31 ; DfO^fl Dan. I, 10. In case of the verbs yy the tri-
consonantal form with repeated second radical appears regularly
in Hebrew in the 3d sing, mas. and fern., and 3d phi. of the
Qal perf. as the trans, form, while the more original biconso-
nantal form is used as the form with intrans. meaning,1 e. g.,
"HV 'make narrow,' but "")¥ 'be nai-row.' Also a considerable
number of regular Pi'el forms from the amplified triconso-
nantal stem — which, as we saw, are practically of non-occurrence
in verbs \'y — are formed from roots yy> e. g.,

It is clear from the foregoing that originally no Pi'el intensive
stem could be formed in the case of the bicousonantal verbs
*\'y and yy < since the Pilel requires three radicals for its
formation. It is not until these roots have been fully assimi-
lated to the triconsonantal form that the Pi^el can be made.
In actual fact the 'Pi'el of verbs *V'j7. as was stated above, does
not properly occur in Hebrew. We find in its stead the Pi^lel.
Also in the verbs mediae geminatae the PWel occurs as the
intensive stem, although some regular Pixels are found, e. g.,
^H , etc.

With regard to the origin of the JPi'lel there have been a
number of different explanations. In general, opinions as to its
origin may be divided into two main currents according as it is
supposed to have arisen independently in the verbs 'V'Jf and
yy ,2 the agreement in final form being then accidental, or it

present, e. g., J>$1D • ^IPf » yV • "IIN (two gutturals !), etc., proving that
if ^ was second radical the presence of a guttural in the root did not
prevent contraction as is supposed. Roots like ^>"\y (^IJ^) can» there-
fore, be explained only as a separate class distinct from the roots \'y ,
We cannot otherwise account for the difference in meaning between two
roots, otherwise identical, like ^y , (tyy Is- 2^' 10) 'act unjustly,
corruptly,' and *y)y (part. plu. fern, ft^jy) ' suckle ' — the former is tri-
consonantal with ^ as middle radical, the latter biconsonantal.

1 Of. Ges.-Kautzsch81, §67a, 1 2; Konig, Lehrgeb., 1, pp. 320, 321.

2 Thus Bottcher, Lehrgebdude, § 1016, § 1030, 2; Olshausen, § 251 b,
252, 254; Konig, Lehrgeb. I, pp. 451 and 349.

Vol. xxvii.] The Pi'lel in Hebrew. 311

arose in the one class and was transferred to the other by anal-
ogy. The latter view, that the Pi^lel arose in one class and
was transferred to the other, is undoubtedly correct whatever
explanation is otherwise adopted. It is the view followed by
Ewald,1 Hartmann,2 Stade,3 and Barth,4 Ewald and Hartmann
believing that the Plllel arose in the verbs ]?"]}, while Stade
and Barth, although differing in other respects in their explan-
ations of the form, believing that it arose in the verbs V'17-

Stade's explanation, that the Pi'lel arose from the Qal stem
qdma by reduplicating the final radical in order to indicate the
intensive stem, producing qdmdma, qdmlm, qomem, is unten-
able especially because the Qal stem is not qdma, but qdma.
The long d becoming 6 in qomem is thus unaccounted for.

Barth's explanation, which is based on the triconsonarital
theory of verbs V'Jf, has been accepted by Kautzsch (Gesenius-
Kautzsch, Hebrew Gram."2', § 72 m), although he follows the
biconsonantal explanation of the verbs V'J? (and yy , §67).
There are, however, certain difficulties in Barth's exposition,
apart from the fact that it is based on the triconsouantal explan-
ation, which render it impossible to accept his theory. His
explanation is briefly as follows.

A weighty indication that the Pl'-lel (Po'lel) did not arise in
the verbs ]}"$ is the fact that they can and do form a regular
Pi1- el, e. g., ^H. etc., in Hebrew as in the other languages.
In the verbs *\"y , Pi'el forms do not properly occur because of
the difficulty of pronouncing an intervocalic [my italics] sharp-
ened waw [as if a doubled 1 could be anything but intervo-
calic !] ; only the Pi'-lel with reduplicated final consonant, as in
DQlp • is found. The reduplication of this final consonant
in this stem of roots V'^ is supported by a similar reduplica-
tion in certain nominal forms in Arabic and Hebrew. In Arabic
no verbal form with this reduplication is made in verbs \'y , —
only nominal forms occur. These nouns, in which no intensive

*' '.°~
meaning is present, are the peculiar infs. like XJ..AAJ ' go away '

1 Lehrbuch*, §§ 121 a, 125a.

2 Op. cit., p. 2,3.

3 Hebrew Gram., § 155 c, d.

4 Die Polel-Conjugation und die Polal-Participien in Semitic Studies
in Honor of Alexander Kohut, Berlin, 1897, p. 83-93, especially p. 84.

312 . L. B. Wolfenson, [1906.

»
med. i), xJj-^y 'be' (^\J mec?. w), etc.; the two infs.

s , ,
' rule ', ialo«-c. ' be pregnant ' ; and the broken plurals

s- ,
«.ft 'pregnant' and JJ>a* 'barren for a long time (camels),'

all from roots med. waw and med. yod. In the case of roots
med. geminatae no corresponding formations occur. In Hebrew

the nouns nfTJ 'pleasure', (V fTJ)> TiT3 'spark' (Arabic
' emit fire ') from roots \'$ have this same reduplication. Only
p!M ' spark ' from WJ ' sparkle ' is from a root JTJ? . This
may be formed on the analogy of its synonym "111*3 . Except
for this single instance, formations in Semitic corresponding to
the Pi'-lel are always from roots *V'J? .

The same result is, according to Barth, arrived at from a con-
sideration of the Hebrew participles D3l£^. t?t7J^. with which
he says DD1"! and the uncertain ^ItJ/ are connected as regards

. T L.L. ' • T

formation. On 77*)^ he lays little stress, since its meaning, and
hence its root, is obscure. These participles are not intensive
in meaning, but are simply Qal. Trans, in force, although
having an apparently pass, or intrans. vowel — in the second
syllable, they are really qattdl forms represented in Hebrew by
D3J > flJD • etc. , in strong roots ; by P"! , "|*¥ in roots med. i.
Accordingly from roots med. u of which no qattdl form occurs
the original form of these participles was qawwdm. A "shar-
pened " 10 being avoided in Hebrew more than a "sharpened"
j [?], the doubled w was given up, being replaced by the redu-
plication of the following radical, so that qawwdm became
qawmdm, qomdm, the change of aw to 6 being similar to that
in W)D from HID •

V T

Like these participles the Pi'lel is to be explained. From
Dip the qittel must originally have been qawwem with a for I
in the first syllable. The intervocalic sharpened w was avoided
by substituting the reduplication of the final radical so ttrat
from qawwem, qawmem becoming qomem arose.

In this explanation there is little that is convincing. That

s> »0^
the nominal forms like XJ^AXJ, etc., have in Arabic a redupli-

cated final consonant like the Pi^lel proves nothing for this ver-
bal stem in Hebrew since the origin of these forms is obscure

Vol. xxvii.] The PWel in Hebrew. 313

and a matter of dispute. Earth's explanation of them is no
better than that of the grammarians which he rejects. He says'

S" >«., S »

that Jo *AXJ , e. g., arose from ^yU , beside which it occurs,
because the phonetic sequence -uiil was unpleasant. This was
obviated by inserting a consonant identical with the last one
after the i so that we get bttinHn. In the first syllable of this
form, $ was changed to «, and thus bdinttnat arose. The changes
that Barth assumes here are all unsupported by similar phenom-
ena elsewhere, and are therefore entirely gratuitous. It may

Ss )o ,,

be that the forms like xJ^-^o arose from an analogical com-

s ° - G »

bination of the two regular infs. like ^^ and ^yj?* • If this

be true there is no organic reduplication in these forms. At

sr , „
any rate there is nothing in forms like XJ..XAJ , about the origin

of which nothing is really known, that is like the Pi'lel, except
the reduplicated final radical. Similarly the isolated forms

fs. ) f s > » ?f '

Jfljs«.£ , J J..AW , JJ»^»- prove nothing.

In Hebrew the nominal forms fllTJ' TITS. TI^J . are entirely
too few to base any conclusions upon. Since ^1^-3 is from a
root y"y and there are only two other examples of this forma-
tion, it is just as possible to conclude that the formation origi-
nated in roots ]?"]} and were transferred to those med. u,
especially since the root of TiT3 , which occurs only in Job.
41, 11, does not occur as a verb in Hebrew.

Likewise the few forms DDl^ , ^Tl V, 001*1 and ^it^ prove

T T ^ T T A

nothing. Barth himself attaches no importance to '?'7}t^ . It
is most likely derived from a root $?"%? . D01"l is usually
explained as having the adverbial ending Q T,cf. Ges.-Kautzsch",
§ 100 g; Gesenius-Buhl14, s. v.* There is no compelling reason
for regarding D011 as anything but an adverb in the three pas-
sages that it occurs. In D011 |D^> Hab. 2, 19 we must connect
00*1*1 with what follows, according to the suggestion made in
the latest (14th) edition of Gesenius' Dictionary. In Lam. 3'
26 it is difficult to see how not to make D011 an adverb. A. V.

1 Die Nominalbildung in den Semitischen Sprachen, Leipzig, 1889,
1891, pp. 211, 212.

2 In the 13th ed. Earth's explanation was given.

314 L. B. Wolfenson, [1906.

translates, "It is good that a man should both hope and wait
quietly ( DDVl) • .etc." In Is. 47, 5, DO11 is clearly an adverb.
Moreover the change of j to 1 in D011 which Barth assumes is
unexplained. This leaves only the two forms DDlt^ and ^l^

upon which to base any conclusions. This would be precarious
even if Barth had correctly explained them. But granting
that they were originally qattdl formations as he says, there is
no support for the supposition that the "intervocalic sharpened
ic," which must originally have been present, e. g., qawwdm,
in verbs " med. w," was any more unpleasant to the ear or dif-
ficult to pronounce in Hebrew in case of forms "med. w"
than in fornfs "med. „/," e. g., ""!*¥ • P"! • We find a consider-
able number of forms with doubled 1 , e. g., ^1^7 < ty$* • "VU^ *
etc., and in verbs v'£ the first radical 1 is regularly doubled in
iheNtph. im-pf:, imv., and inf., e. g., J/1V > etc. In the case of
forms like *Y)^. fVlJ7» etc., Barth tries to obviate this difficulty
by the remark1 " Wurzeln mit durchweg cons, behandelten w,
. . . , gehoren nicht hierher." There is, however, no reason
why "die Wurzeln mit . . . cons. . . . w gehoren nicht hier-
her." There can be no difference between original w in qaw-
tcdm, if such there was (which the biconsonantal theory denies ;
cons. 1 that appears in roots 'V'^ is secondary), and that in
'Tl^. ni^> l"Tn> etc., where 1 appears everywhere as a conso-
nant, cf. Konig, Lehrgebdude I, p. 453. The case of the Nipfts.
like J^"1V Barth does not consider. Moreover even though a

doubled w were objectionable in Hebrew, as the preceding
shows it was not, there is no parallel for reduplicating a radical
in compensation for the lack of doubling in another. The
approved method of compensation for the omission of doubling
is to heighten the short vowel preceding the doubled consonant,
as is done in countless instances in the case of the article, the
Niph. impf., imv., and inf. of verbs primae gutt. , e. g., "IDJ?' .
"?DJ7n > etc. Accordingly it is impossible to assume that an
original qawwdm became qawmem.

Similarly the Pi'lel stem cannot be explained as coming from
an original q&wwem -(qlwwem) becoming qomem. In fact it is

1 Die Pdlel-Conjugation, p. 90, n. 3.

Vol. xxvii.] The Pi'lel in Hebrew. 315

absolutely impossible to assume such a verbal form as qawwem
in roots *V'^ , as it was shown above that these roots had no
middle radical. Consequently no form like qdwwem, etc., ever
existed in Hebrew; the forms "T)#.D'P' etc., with doubled 1
and ' , are not Hebrew.

We must, therefore, look for the origin of the Pi^lel else-
where, and it is in the verbs yy that it is found, as Ewald and
Hartmann saw. It is not necessary to assume with Hartmann
that the Pi'lel represents the III form of the Arabic. Ewald
explained the 6 in DDlD- e- g-> as due to the obscuration of a1.
This d arose from d in compensation for the difficult doubling
of the second radical in D3D • To this explanation Barth objects
that it presupposes an d after the first radical of the root, which
does not occur in the Hebrew period, the form being always
sibbeb, although he assumes d for I in his own explanation of
q&wwem for qiwwem. See above, p. 312, ^[ 2.

In the imperfect, however, as well as the forms agreeing with
it in structure, viz., the imv. and inf., the regular forms are
D3D* > etc. Here the original d after the first radical is retained
throughout. iesdbbeb accordingly would become iesdbeb, with d
lengthened from d in compensation, as soon as the doubling of
the second radical is given up. The d is then obscured to 6, as
frequently in Hebrew, and hence the form ^DTD'- From the
impf. the 6 was then transferred to the perf., e. g., D^D-2

That there is a tendency to avoid the occurrence of three
identical consonants in two successive syllables as in the Pi'-el
of verbs ^"Jf, not only in Hebrew but in other languages, is seen
from the fact that in classical Arabic beside such forms as

c>-«3-«ift J" , (jd^aiiJ', etc., with doubled 2d radical identi-

- o *- *• *" *:'

cal with the 3d, we find oocXx , ^^OA*, ^xiai', etc., with the
3d radical replaced by the diphthong di, on account of the

1 So also Konig, Lehrgeb. I, p. 349, in the case of verbs V" V ; Bickell,
§ 116 ; and Land § 55 (two latter quoted by Konig).

* In this explanation I have followed the. principles established by
Prof. Haupt, viz., that the impf. is older than the perf. (cf. his article
in Jour. Royal Asiatic Soc., New Series X, 1878, pp. 244-252), and that
the origin of verbal forms is to be sought in the impf. as the more
original form.

316 Wolf enson, The. Pi'lel in Hebrew. [1906.

" heaping up " of consonants. Cf. Fleischer, Kleinere Schriften
I, p. 138; Wright-deGoeje, Arabic Gram.3 I, p. 69 C. In mod-

«- * •" ^
ern Arabic even the simple Qal forms like v^i>tXx> with repeated

consonant are given up, and forms like c^otXx> only are used.1
In fact modern Arabic goes even farther in the case of such
forms, using the form of verbs tertiae ^ instead of those

mediae geminatae,* e. g., oojuo for

The Pi'-lel is therefore formed on the basis of the Pi^el of
verbs p''^; the doubling of the second radical is given up on
account of the tendency to avoid a succession of three identical
consonants in two successive syllables, and the preceding short
a is lengthened in compensation to «, and this is further obscured
to 6. The corresponding passive form, the Pi'ldl, has a in the
second syllable, e. g., 331D*> 331D- The indication of the dis-

tinction between act. and pass, by i (em Hebrew) and a,
respectively, is regular in Arabic in the impf., not only of the
intensive stems II, III, but also of the IV, VII, VIII, and X

forms, e. g., II form *&b act., JJAJpass., etc.

From the verbs the Pillel was transferred to the verbs

1 Cf. Spitta, Gram, des arab. Vulgardialectes von Aegypten, Leipzig,
1880, p. 216.

Contributions to Comparative Philippine Grammar* — By
FRANK R. BLAKE, Ph.D., Johns Hopkins University,
Baltimore, Md.

INTRODUCTION.

The languages of the Philippine Islands so far as they are
known2 form a closely related group of tongues belonging to
the great Malayo-Polynesian family, which embraces the lan-
guages of practically all the islands between the east coast of Asia
and the west coast of America south of 30° north latitude,
with the exception of Papua and Australia, and also includes
the languages spoken on the Malay peninsula at the south-east-
ern extremity of the continent of Asia, and on the island of
Madagascar off the coast of Africa.

1 In the spelling of Philippine words in this paper, k is used instead
of the older c and qu ; g instead of gu before i, e; w instead of conso-
nantal u before a vowel ; i for initial y before a consonant ; but with
these exceptions it has seemed best to retain the traditional orthography.
As in most of the languages the accent of words is not given, the accent
marks have been omitted throughout, except where they indicate a dif-
ference in meaning, and in the case of ^ and x in Tagalog, which indi-
cate a final glottal catch (cf. below, p. 335). For the reformed spelling
in Tagalog, cf . the foot-notes to my paper, The Expression of Case by
the Verb in Tagalog, in this volume of the Journal.

2 The languages of the Negritos, the diminutive black people who
live in scattered tribes in the interior mountain ranges of the larger
islands, and who are probably the remnant of the aborigines of the
Archipelago, are said by certain Spanish authorities (cf . El Archipielago
Filipino, Washington, 1900, vol, 1, p. 229; Lacalle y Sanchez, Tierras y
razas del archipielago filipino, Manila, 1886, p. 246) to be entirely dif-
ferent from those of the other inhabitants of the islands. These author-
ities state furthermore that the idioms of all the Negritos practically
constitute one language, and that this language is of monsyllabic struc-
ture as opposed to the dissyllabic structure of the Malay tongues. How-
ever true this may have been originally, at the present day it is certain
that those Negrito dialects about which anything is known are very
similar in vocabulary and grammatical structure to the other Philip-
pine languages. It is probable, however, that the Negritos have in such
cases more or less completely adopted the languages of the neighboring
Malay tribes (cf. A. B. Meyer, Die Philippinen, II. Negritos, Dresden,
1893 (Konigl. ethnogr. Museum zu Dresden, ix), p. 36 ff .

318 F. R. Blake, [1906.

It is perhaps useless to hazard any conjectures as to the prim-
itive seat of the Malayo-Polynesians, whose ancient history is
practically a sealed book, but it seems most likely that the
cradle of the race was on some of the numerous islands which
it now inhabits, possibly some of the large islands in the vicin-
ity of the Malay peninsula.

From this birthplace the ancient Malayo-Polynesians, forced
doubtless by the increase in population, must have spread out in
a series of waves or swarms, just as in the case of the primitive
Indo-Europeans and Semites. Crossing at first by means of
their canoes over the comparatively short stretches of sea between
their home and the neighboring islands, they gradually pushed
further and further out into the unknown, passing from island
to island until they had occupied almost all the available land
space of the Pacific. Some of the islands they probably found
unoccupied, in others they must have come in contact with an
inferior black race similar to that inhabiting Papua and Australia,
as is shown by the remnants of this race which are found pushed
back into the interior mountain ranges of some of the larger
islands, notably the Philippines.

The peopling of the Malayo-Polynesian territory probably
took place in three great waves or series of waves, to which
correspond the three grand divisions of tha Malayo-Polynesian
languages, viz., the Polynesian, the Melanesian, and the Malay.
The 180th meridian forms approximately the boundary between
the Polynesian and Melanesian divisions from the north as far
south as the latitude of the Fiji islands, practically all the lan-
guages spoken east of this line being Polynesian. Further south
the line bends to the west, the native language of New Zealand
belonging to the Polynesian division. West of the dividing
line the Melanesian division extends in a north-westerly direc-
tion from the Fiji islands on the south, including the languages
of the principal islands of Melanesia and Micronesia. The
Malay embraces the languages of the Malay peninsula, the East
India islands, the Philippines, and Madagascar.

The Philippine languages, then, may be more accurately
defined as a subdivision of the Malay branch of the Malayo-
Polynesian family of speech.

The estimated number of Philippine languages varies accord-
ing to the different authorities. The well known Spanish Fili-

Vol. xxvii.] Contributions to Philippine Grammar. 319

pinologist W. E. Retana, in his latest bibliographical work on
the Philippines,1 enumerates twenty-five different idioms; the
great Philippine specialist, Professor Ferdinand Blumentritt,
of Leitmeritz in Bohemia, in his brief survey of Philippine races
and languages,2 mentions at least thirty; while in an encyclo-
pediac work on the Philippines prepared by the Jesuits, 'El
Archipielago Filipino,'3 the number given exceeds fifty.

Of many of the languages enumerated in the larger estimates,
practically nothing is known but the name, and it is quite pos-
sible that many of these names are simply alternate designations
of the better known languages, or, at most, designations of
some slightly variant dialect. Beginning at the extreme north
of the Archipelago, the languages about which anything definite
is known are as follows.

Batan is the language of the Batan and Babuyan islands to
the north of Luzon.

On the island of Luzon, Tagalog, the most important and
best known of the Philippine languages, is spoken from coast
to coast, in the middle region of. the island, in the latitude of
Manila Bay. On the west coast its territory does not extend
north of the Bay, but on the east it reaches as far north as the
province of Isabela, the most northerly but one of the provinces
on the east coast, in which is situated the town of Palanan,
where Aguinaldo was captured by General Funston. On the
south and south-east it extends some distance down into what
might be called the tail of Luzon, trenching on the domain of
Bikol, which occupies the remainder of the southern part of the
island. In the region north and west of the Tagalog district
are spoken a number of languages. Ibanag prevails in the north-
east, in the valley of the Cagayan river, the greatest tobacco-
raising district in the island; Ilokan occupies the north and west
coasts, extending as far south as the gulf of Lingayen, between
which and the Bay of Manila are found Pangasinan, Tino or
Zambal, and Pampangan. In the mountainous district of the
interior are spoken the various Igorot dialects, among which

1 Catdlago abreviado de la biblioteca filipina, Madrid, 1898.

2 Cf. List of Native Tribes of the Philippines and of the languages
spoken by them, trans, by O. T. Mason in Report of Smithsonian Inst.
for the year ending June, 1899; Washington, 1901, pp. 527-547.

3 Washington, 1900 ; cf. vol. 1, pp. 1-148 passim.

320 F. R. Blake, [1906.

it is probable that Gaddan, Ginaan, Ilongot, and Isinay are to
be classed.1

On the Bisayan islands, which lie between Luzon and Min-
danao, and on the north and east coasts of the latter island,
Bisayan is spoken in a number of different dialects.2

Sulu is used by the Moros of the Sulu subarchipelago, which
extends from the western exti'emity of Mindanao towards
Borneo. The Moro tribes of Mindanao, which occupy parts of
the west and southwest of that island, speak two almost identi-
cal dialects, Magindanao and Malanao. Of the numerous other
idioms reported as spoken on Mindanao, we know practically
nothing about any except Bagobo, which is found near the
great volcano Apo in the south-eastern part of the island, and
Tiruray, which occupies a district near the Moro territory in
the south-west.

The island of Mindoro, which lies to the south of Luzon and
west of the Bisayas, forms the domain of Mangyan, about
which, so far as I know, nothing has yet been published.1 The
principal language of the Calamianes and the long narrow island
of Palawan, which form a chain stretching from Mindoro to the
south-west towards Borneo, is Tagbanwa, of which the idioms
of Agutaya and Cuyo, two small islands between Palawan and
the Bisayas, are perhaps simply dialects,2 though they are usually
given as distinct languages.

The tribes that speak these languages fall into three general
groups according to their religion. Those that speak Batan,
Ibanag, Ilokan, Pangasinan, Zambal, Pampangan, Tagalog,
Bikol, Bisayan, Agutayan,2 and Koyuvan3 are Christians; the
Sulus, Magindanaos, and Malanaos are Mohammedans; while
the remaining tribes mentioned are still pagans.

Three different alphabets are in use in the Archipelago, viz.
1) that of the pagan Tagbanwas and Mangyans;4 2) that of the

1 Cf . Report of the Philippine Commission, 1903, Part 2, p. 780.

* Cf. ray paper The Bisayan Dialects, JAOS. xxvi, 1905, pp. 120-136.

3 Cf. Blumentritt, The Philippines, trans, by D. J. Doherty, Chicago,
1900, p. 24.

4 The Mangyan alphabet, however, is treated in the following, viz.,
Blumentritt, Die Mangianenschrift von Mindoro, Braunschweig, 1896;
A. B. Meyer, Schadenberg and Foy, Die Mangianenschrift von Mindoro,
Berlin, 1895=^l&7iancK. u. Berichte d. Konigl. zoolog. u. anthropol.-
ethnogr, Museums zu Dresden, No. 15.

Vol. xxvii.] Contributions to Philippine Grammar. 321

Mohammedan tribes, the Arabic alphabet with some additional
signs to denote some peculiar native sounds;1 3) the Roman
alphabet introduced by the Spaniards, in which all the languages
of the Christian tribes, and all those of pagan tribes which have
been reduced to writing by Spanish missionaries, are written.

At the time of the Spanish conquest, the principal of the now
Christian tribes possessed alphabets that are practically identi-
cal with those of the Tagbanwas and Mangyans," and it is
probable that the Mohammedan tribes originally had similar
alphabets. These ancient alphabets have in both cases been super-
seded by that of the race whose religion has been adopted. In
the Mohammedan tribes no trace of them has been preserved,
and their use in the Christian tribes seems to have died out
about the middle of the eighteenth century.3

The archetype of these natives alphabets seems to have been
of Indian origin. As in the Indian alphabets, every consonan-
tal character without addition represents the consonant fol-
lowed by the vowel «, the other vowels being indicated by
secondary marks. There was no way of expressing a consonant
without a following vowel, hence such a consonant was omitted
in writing.4

1 Similarly there are some additional characters in the Arabic alpha-
bet adapted to Malay, Persian, Turkish, and Hindustani, cf. Marsden,
A Gram, of the Malayan Language, London, 1812, p. 1 f. ; Salemannund
Shukovski, Persische Gram. Berlin, 1889, § 1 ; A. Miiller, Turkische
Gram. Berlin, 1889, § 5; Vinson, Manuel de la langue hindoustani, Paris
1899, p. 5. In like manner the Amharic alphabet is the Ethiopic with
some additional characters to denote some peculiar Amharic sounds,
while Coptic is written in the Greek uncial alphabet with seven addi-
tional characters borrowed from Demotic: cf. Praetorius, Die Amharische
Sprache, Halle, 1879, p. 17, § 1 a ; Steindorff , Koptische Grammatik,
2te ed., Berlin, 1904, p. 5.

2 Cf. Marcillo y Martin, Estudio de los antiguos alfabetos filipinos,
Malabon, 1895.

3 Totanes in his Arte de la lengua tagala, Sampaloc, 1745, states that
at his time very few natives could read this alphabet, and that hardly
any could write it : cf. the later edition, Binondo, 1865, p. 1.

4 This defect was remedied by the Austin friar Francisco Lopez, who
in his Ilokan catechism (1631) written in Tagalog characters made use
of a diacritical mark, similar in its nature to the Sanskrit virama or
Arabic sukun, to indicate a consonant standing alone; cf. El Archipielago
Filipino, vol. 1, p. 227.

VOL. xxvii. 22

322 F. ft. Blake, [1906.

The Roman alphabet, which is used in writing the native lan-
guages, was formerly conformed to the peculiarities of Spanish
orthography, but lately a number of improvements in spelling
have been introduced, the most important being the use of k
for c and qu, and w for consonantal y»

The Philippine languages have been influenced to some extent,
principally in their vocabulary, by the languages with which
they have come in contact. The vocabularies of some of them,
notably Tagalog and Bisayan, contain, in common with the
other Malayan languages, a number of Sanskrit words, e. g.,
Tagalog and Bisayan basa 'read' (San. bhasa 'languages'),
halaga 'price' (San. argha).1

The languages spoken by the Mohammedan tribes, the most
important of which are Magindanao and Sulu, contain a number
of Arabic words, e. g., Magindanao and Sulu dunia 'world'
(Arabic L*J4> dunid], Mag. alatala, alahutaala, Sulu allah-

taala ' God ' (Arabic JLxi* &1J! allahu ta'-dld ' God, may he be

(5
exalted').

The Christian tribes have adopted a considerable body of
Spanish terms, e. g., Dios ' God,' pade, pare ' priest.'

A few Chinese words are found in Tagalog as designations
of things specifically Chinese, e. g., cha or sa 'tea' (Chin.
tcha), miki 'a kind of vermicelli' (Chin, mi-ki}.

At present the various languages are being subjected to the
influence of English, and English words will probably be more
or less extensively borrowed. Already in the northern part of
Luzon the English phrase ' no got ' is in common use.4

Spanish, besides influencing to some degree the native vocab-
ularies, has also left its mark in a few cases on the grammatical
construction. In Tagalog, for example, the cumbersome native
method of coordinating pronouns and numerals, as in kami niya
'he and I' literally 'his we,' has been more or less completely

1 Cf. H. Kern, Sanskritische woorden in het Tagala, Bijdragen tot de
Taal-, Land- en Volkenkunde van Nederlandsch Indie, 4de Volg., D.
4, 1880, pp. 535-564; Sanskritische woorden in het Bisaya, ibid., 4de Volg.,
D. 5, pp. 128-135; T. H. Pardo de Tavera, El Sanscrito en la lengua
agala, Paris, 1887: cf. also my paper, Sanskrit Loan-words in Tagalog,
JHU. Circs., No. 163, pp. 63-65.

s Cf. A. E. Jenks, The Bontoc Igorot, Manila, 1905 (=Ethnological
Survey Publications, vol. 1). p. 158.

Vol. xxvii.] Contributions to Philippine Grammar. 323

driven from the field by the simpler Spanish construction with
copulative conjunction.1

The materials for the study of the Philippine languages con-
sist of texts, collections of conversational phrases, grammars,
dictionaries, and vocabularies. Grammars and dictionaries of
some sort exist of most of the languages mentioned; the others
must be studied without these helps. The languages that are
included in the following comparative studies are, viz., Tagalog,
Bisayan (Cebuan, Hiliguayna, Samaro-Leytean, Harayan), Bikol,
Pampangan, Pangasinan, Ilokan, Igorot (Nabaloi, Bontok),
Ibanag, Batan, Magindanao, Sulu, and Bagobo.2

GENERAL FEATURES.

The most important characteristics which the Philippine lan-
guages possess in common are the following.

Words are made up of roots and particles. Roots are mainly
dissyllabic and indicate nominal or verbal ideas; practically all

1 Cf. W. G. Seiple, The Tagalog Numerals, JHU. Circs., No. 168, pp.
79-81.

s The principal grammars and dictionaries employed are, viz. : Totanes,
Arte de la lengua tagala, reimpr., Binondo, 1865; Campomanes, Lec-
ciones de gramdtica hispano-tagala, Manila, 1872 ; Minguella, Ensayo
de gramatica hispano-tagala, Manila, 1878 ; Noceda, Vocabulario de
la lengua tagala, 3aed., Manila, 1860; Zueco, Metodo del Dr. Ollendorff
.... adaptado al bisaya, Manila, 1871 : Bermejo, Arte conpendiado
de la lengua cebuana, 2a ed., Tauibobong, 1894 ; Mentrida and Aparicio,
Arte de la lengua bisaya-hiligayna, Tambobong, 1894 ; Lozano, Cursos
de lengua panayana, Manila, 1876 ; Figueroa, Arte del idioma visaya
de Samar y Leyte, 2aed., Binondo, 1872; Encarnacion, Diccionario
bisaya-espanol, 3a ed. , Manila, 1885 ; San Augustin and Crespo, Arte
de la lengua bicol, Manila, 1879 ; Bergano, Arte de la lengua pampanga,
2aed. (?), Sampaloc, 1736; Pellicer, Arte de la lengua pangasinana,
reimpr., Manila. 1862; Cosgaya, Diccionario pangasinan-espanol, Ma-
nila, 1865 ; Naves, Gramdtica hispano-ilocana, 2a ed., Tambobong,
1892; Carro, Vocabulario iloco-espanol. 2aed., Manila, 1888; Sheerer,
The Nabaloi dialect, Ethnological Survey Publications, vol. II, Part II,
pp. 95-171, Manila, 1905 : Jenks, The Bontoc Igorot, vol. 1 of series cited,
Manila, 1905, pp, 227-248 ; De Cuevas, Arte nuevo de la lengua ybandg,
2aed., Manila, 1854; (Batan grammar) cf. Retana, Archivo del bibliofilo
iilipina, vol. II, Madrid, 1896, pp. xxxviii-xl ; Juanmarti, Gramdtica
de la lengua de Magindanao, Manila, 1892 ; Cowie, English-Sulu-Malay
Vocabulary, London, 1893 ; Gisbert, Diccionario bagobo-espanol and
espanol-bagobo, Manila, 1892.

324 F. R. Blake, [1906.

may be used unchanged as significant words; e. g., Tagalog
tawo 'man,' ibig 'wish, want.' Particles are mainly monosyl-
labic, some being independent words indicating pronominal and
adverbial ideas, others being used only in combination with
roots to form derivative nouns and verbs, e. g., Tagalog ka
4 thou,' net 'now, already,' mag, a prefix used to form active
verbs, e. g., mag-laro 'play, sport' from lard. From these
ultimate components of the language other words are formed :

a) by reduplication of the root, e. g., Tagalog susulat 'will

write ' from sulat ' write ;'

b) by the combination of two or more particles, e. g., Tagalog

na-man ' also,' from na ' now ' and man ' even;'

c) by the combination of the root with one or more derivative

particles, e. g., Tagalog s-um-ulat 'write (imper. and
inf.), s-in-ulat-an 'was written on,' from sulat.

These languages are practically non-inflectional, there being
no inflection except in some few instances in the pronoun and
the verb, the variation being regularly at the beginning of the
word. Pronouns are varied to express case, as a general thing
three cases being distinguished, a nominative, a so-called geni-
tive that is also the case of the agent and instrument, a so-called
oblique that is used to represent all locative relations, place
where, place whither, and place whence, including the dative
and ablative of persons; e. g., Tagalog ito 'this,' nito 'of
this,' dito ' in or to this.' In the verb differences in voice, mode
and tense may be indicated by change of the initial sound of a
form, e. g., Tagalog mag-laro 'to play,' naglaro 'played,'
paglaro, passive stem of same verb, where m indicates the infini-
tive, n the preterit, andjo the passive.

There is no formal distinction of gender even in the case of
the pronouns of the third person. Whenever it is necessary to
indicate the gender expressly, words meaning 'male' and 'female'
must be used in connection with the epicene noun or pronoun,
e. g., from Tagalog kabayo ' horse,' kabayo-ng lalaki 'stallion,'
and kabayo-ng babayi ' mare ;' except in the case of certain
nouns of relationship, where different words are used to express
difference in gender, e. g., Tagalog ama 'father,' ina 'mother.'
Even with these nouns the same word often denotes a relative
of either sex, and the words for ' male ' and ' female ' must be

Vol. xxvii.] Contributions to Philippine Grammar. 325

used when it is necessary to avoid ambiguity, e. g., Tagalog
anak 'son or daughter,' anak na lalaki 'son,' anak na babayi
' daughter.'1

These languages possess what might be called personal articles,
i. e., words of a particular nature which are placed before names
of persons to denote case, e. g., Tagalog si Pedro 'Pedro,' ni
Pedro 'of Pedro.'2 Many of them also have an article, the so-
called inclusive article, which is placed before the name of a
person to denote that with him are included those who are asso-
ciated with him in any way, as his companions, friends, family,
etc., e. g., Tagalog sina Pedro 'Pedro and his associates.'

The pronoun of the first person plural has regularly two forms,
one of which includes, while the other excludes, the person
addressed; for example, a Tagalog might say to a Spainard
tayo-ng kristiano ' we Christians,' using the inclusive ' we ' tayo,
but kami-ng Tagalog ' we Tagalogs,' using the exclusive ' we '
kami. 3

An extensive use is made of certain particles called ligatures
to connect words, phrases, and sentences which stand to one
another in the relation of modifier and modified.4 They stand for
example between adjective and noun; verb and adverb; noun
or verb ajid dependent clause, in this case playing the part

1 These remarks apply also to the expression of gender in Malay, (cf .
Marsden, op. cit. p. 29 ; Crawfurd, A Gram, and Diet, of the Malay
Lang., London, 1852, vol. i, p. 10 ; Seidel, Praktische Gram, der Malay
ischen Sprache, Hartleben's Verlag [Wien, Pest, Leipzig], p. 18); and in
Javanese (cf . Bohatta, Praktische Gram, der Javanischen Sprache, Hart-
leben's Verl.), p. 31.

2Personal articles are found also in Malay and Madagascan ; in Malay
it is si, cf. Favre, Grammaire de la langue malaise, Vienne, 1876, pp.
59, 92. The principal ones in Madagascan are i and ra, cf. Brandstet-
ter, Tagalen und Madagassen, Luzern, 1902, p. 79 f.; Parker, A concise
Gram, of the Malagasy Language, London, 1883, p. 47.

3 The same distinction is made also in Malay between kita (inc.) and
kami (exc.), and in Madagascan between isika (inc.) and izahay (exc.);
cf. Marsden, op. cit., p. 45 ; Parker, op. cit., p. 39.
• 4 Traces of similar particles are found in Madagascan and some other
Malayan languages, cf. Brandstetter, Tag. u. Mad., pp. 76, 83. In lan-
guages of other families, similar particles are the connective i in mod-
ern Persian (cf. Satemann and Shukovski, op. cit., p. 30, § 16), and the
genitive sign n in Egyptian and Coptic (cf. Ermann, Agyptische Gram.
2te Aufl., Berlin, 1902, p. 64; Steindorff, op. cit., pp. 79 f., 83).

326 F. R. Blake, [1906.

of relative pronouns and subordinate conjunctions; e. g. , in
Tagalog tawo-ng mabuti 'good man,' mabuti-ng ginaicd 'well
done,' tawo-ng minamahal nang lahat ' a man who is esteemed
by all,' the modifier and modified are connected by the liga-
ture -ng.

The ideas of ' to be in a place ' and ' to have ' are not expressed
by verbs, but by particles which may be called quasi-verbs, in
Tagalog na, may, e. g., ang library na sa lamesa 'the book is
on the table;' ito-ng tawa'y may asawa 'this man has a wife.'

Verbs are generally made by combining derivative particles
with the root. These particles are very numerous and their
uses very various. By means of them voice, mode, and tense
are distinguished, and also a variety of other modifiations of the
verbal meaning, such as the causative, emphatic, etc. ; for exam-
ple from a root aral in Tagalog are formed,

um-aral ' teach.' magsi-aral 'teach (of many).'

mag-aral ' study, learn.' magpaka-aral 'teach earnestly.'

mang-aral 'preach.' ungm-aral 'taught.'

maka-aral ' be able to teach.' nag-aral ' learned.'
magpa-arul 'order, command aralin 'be taught.'

to teach.' inaral ' was taught.'

maki-aral 'join with someone pag-aral-in 'be studied.'
in teaching.' pinag-aral 'was studied.'

pa-aral 'ask for instruction.' etc., etc.

Perhaps the most salient feature of these languages is the pre-
vailing use of the passive construction, active verbs not being
used except when the agent is the most emphatic element of
the sentence; for example in Tagalog in the sentence 'he is
reading a book,' 'he' is more emphatic than the indefinite ' book,'
hence the active is used, viz., siya'y bungmabasa nang libra,
while in the sentence ' he is reading this book ' the definite
object is ordinarily more emphatic than the agent, hence the
passive is employed, e. g., ito-ng libra' y binabasa niya.*

In the present imperfect state of our knowledge of the Philip-
pine languages any complete classification and subdivision is of
course impracticable, but it is possible nevertheless to distinguish
certain general groups.

1 Of. my paper, The Expression of Case by the Verb in Tagalog, in
this volume of the Journal, pp. 18&-189.

Vol. xxvii.] Contributions to Philippine Grammar. 327

The most important of these is a Northern Group, including
the principal languages of North Luzon and the islands to the
north, viz., Batan, Ibanag, Ilokan, Pangasinan, and the Igorot
dialects as far as they are known; and a Central Group, includ-
ing Tagalog, Bikol, and Bisayan. Between these two groups
lies Pampangan, which partakes in a measure of the peculiarities
of both. In the south the dialects of the Magindanao and
Malanao Moros belong together, while Bagobo and Sulu occupy
isolated positions, Sulu being more like Malay than any of the
other Philippine languages. The principal phonetic difference
between the languages of the Northern and Central Groups is
that the former have no h sound (cf. below, p. 335).

The principal differences in grammatical structure between
these two groups are the following.

In general the languages of the Northern Group form their
plural by reduplication of the singular, e. g., Ilok. balay
' house,' balbalay ' houses.' The languages of the Central Group
employ a special prepositive particle manga for this purpose,
e. g., Tag. bahay 'house,' manga bahay 'houses.'

The Northern Group possesses the ligature a which is not
found as such in the Central Group, e. g., Ilok. naimbag a tao
' good man.'

In the Northern Group the personal pronouns have in general
two forms of the nominative, an emphatic and a non-emphatic
form ; for example, ' I ' in Ilokan is repi'esented by siak when
emphatic, by ak when not emphatic. In the Central Group as
a regular thing only one form of the nominative is employed,
e. g., Tag. ako 'I.' On the other hand, in the Central Group
these pronouns have both a prepositive and a postpositive form
of the genitive, e. g., Tag. aking ulo and ido ko 'my head;'
while in the Northern Group only postpositive forms are found,
e. g., Ilok. balay ko 'my house,' aso-k 'my dog.'

As a general thing the languages of the Northern Group dis-
tinguish only two tense forms of the verb, viz., a preterit and
a form to represent all the other tenses, e. g., Ilok., mangara-
mid, pret. nangaramid 'do, make.' In the Central Group as
many as four tense and mode forms are distinguished, viz., a
modal, used for infinitive, imperative, and subjunctive; a future;
a preterit; and a present: e. g., Tag.,
' maglaro ' to play.' naglaro ' did play.'

maglalaro ' will play.' naglalaro ' is playing.'

328 F. R. Blake, [1906.

The languages of the Central Group possess three passive
forms, by means of which not only the direct object of the
verbal action may be made subject, but also words standing in
other case relations of the verb, such as dative, instrumental,
ablative, etc.1

In the Northern Group this function of the verb has been still
further developed, the languages of this group possessing not
only three passives, but also a number of other passive verbal
forms called verbals, made on the basis of the verbal noun of
action with prefixed pag, pan, etc. These take the place to
some extent of the i and an passives of the Central Group.
Examples of these forms in Ilokan are,
daytoy ti pagpunas-mo kadagiti pinggan ' wash the plates with

this (this the washing-instrument thine to the plates).'
ania nga oras ti panagmisa ti pare ' at what hour does the priest
celebrate mass (what hour the mass-celebrating-time of the
priest)?'

asin ti pagsurataiyo (<*£m-yo)* 'to whom are you writing (who
the writing-aim yours)?'

Pampangan, which, as was stated above, partakes in a meas-
ure of the peculiarities of both the Northern and Central Groups,
agrees with the Northern Group in not having the consonant
h, and in possessing the ligature a, both emphatic and non-
emphatic forms of the nominative, and exclusively postpositive
genitive forms of the personal pronouns ; but with the Central
in possessing more than two tense and mode forms of the verb,
viz., future, present, and preterit, and in the absence of the
verbals which are so characteristic of the languages of the
Northern Group. The plural of the noun is indicated by a
special form of the article as often in Ilokan, e. g., ing tawo
' the man,' ding tawo ' the men,' except in the vocative, where
it is denoted by prepositive manga, e. g. , manga tawo, as in
the Central Group. A special characteristic of Pampangan is
the large number of forms which the personal pronouns have in
the genitive, e. g. , under various conditions 'mine' is ko, koo,
ke, kee, or da; 'his,' no, noo, ne, nee, or na.

1 Cf. my paper The Express, of Case by the Verb, cited above.

2 In Ilokan the passive endings en and an + the genitive yo ' you (pi.)'
give eiyo, aiyo. The phonetics of the process are not entirely clear.
The spelling may represent an assimilation of n to y, or perhaps a nasal-
ization of the vowels e and a.

Vol. xxvii.] Contributions to Philippine Grammar.

329

Magindanao seems to be more closely related to the Central
Group than to the Northern, forming the plural of its nouns
with manga, arid making three tense and mode forms, viz.,
present, preterite, and imperative, e. g. , sumulat 'writes,'
sinumulat 'wrote,' sulat or panulat 'write (imperative).' It
is, however, apparently without A, and possesses the ligature a
like the languages of the Northern Group. The most charac-
teristic peculiarity of this language are the forms of the per-
sonal pronouns with prefixed or infixed I element (cf. below,
p. 372), e. g., laki, salaki 'my,' salkitanu, lekitanu 'we.'

Sulu, like Malay, possesses in the declension of its noun no
special plural form or plural particle, and forms its tense by
means of auxiliaries, e. g. , tog na aku 'I sleep,' bakas aku
matog ' I have slept.' The ligature, which is so characteristic
of all the other Philippine languages, is here comparatively lit-
tle used.

NOTES ON PHONOLOGY.1
Original Philippine Sounds.

From a comparison of representative words in the various

languages it is evident that the primitive Philippine language

possessed the following sounds, viz.,

vowels a, i, u or o Palatals \

Labials p, b, m, w and >- &, g, y

Dentals £, d, n, I, s Gutturals ) ng*

Cf. the following comparative list of words in the principal

Philippine languages.

' road '

' moon '

' drink '

' cook '

'five'

' eight '

Tag.
Bis.

daan
dalan

buwan
bulan

inum
inum

luto
luto

lima
lima

walo
walo

Bik.

dalan

bulan

inum

luto

lima

walo

Pamp.
Pang.
Ilok.

dalan
dalan
dalan

bulan
bulan
bulan

inum
inum
inum

luto
luto
luto

lima
lima
lima

walo
walo
walo

Iban.

dalan

tulan

inum

luto

lima

walu

Mag.
Sulu

lalan
dan

bulan

inum
hinom

luto
lutoh

lima
lima

walu
walu

1 In general the vowels of Philippine words have the Italian pronun-
ciation; the consonants are to be pronounced as in English unless other-
wise stated.

- Written ng when it stands at the beginning of a syllable.

330 F. E. Slake, [1906.

per. art. ' sickness ' ' eat ' ' sail ' ' heaven '

Tag. si sakit kain layag langit

Bis. si sakit kaon layag langit

Bik. si kan lauag langit

Pamp. sakit kan layag

Pang. si sakit kan layag

Hok. si sakit kan layag langit

Iban. si takit kan layag langit

Mag. si sakit kan layag langit

Sulu sakit layag

The diphthongs ai and au also probably formed part of the
primitive phonetic system ; ai is usually written ay, ai, and au,
ao or au-, cf.

' lake '
danao

' house '

'day'

Tag.

bahay

arao

Bis.

balay

adlao

Bik.

aldao

Pamp.

balay

Pang.

baley1

ageo1

Ilok.

balay

aldao

Iban.

balay

aggau

Mag.

walay

Sulu

bai

adlau

danao

lanau

J lanau
( danau

To these sounds are probably to be added,

a) an indistinct vowel e (cf. below, p. 331 f.),

b) r (cf. below, p. 333),

c) the glottal catch (cf. below, p. 335).
and perhaps h (cf. below, p. 335 f.).

Remarks on the Vowels.

The letters o and u are constantly interchanged in the same
language in writing, e. g. , Tag. otang, utang 'debt,' and
according to the statements of the grammars this graphic dif-
ference is represented by a corresponding difference in the pro-
nunciation. Hence it is hardly possible to distinguish between
primitive Philippine u and o.

1 For the significance of this spelling with e, cf. below, p. 331.

Vol. xxvii.] Contributions to Philippine Grammar. 331

There is also a similar interchange between the letters i and e,
and the sounds represented by them, e. g., Tag. babayi, babaye
' woman,' but this interchange is so limited in scope, the e cor-
responding to i rarely occurring, that all such cases are most
conveniently referred to primitive Philippine i.1

The letters e and o are sometimes used to indicate the con-
traction of the diphthongs ai and au, but usually in these
cases the origin of the e or o is perfectly clear, and there is no
chance of confusion with the other e and o. In the Central
Group this contraction seems to be sporadic and confined to the
final syllable of certain very common words, e. g., Tag. and Bis.
tney 'have,' Bik. dey ' not to have,' where the writing ey indi-
cates the pronunciation e; Tag. ikao^y for ikao ay, where the
elision of the a of the particle ay points to the pronunciation
ikd. In Ibanag and Pampangan these diphthongs at the end of
a word are regularly contracted before a possessive suffix begin-
ning with a consonant, e. g., Iban. bale-k 'my house' Pamp.
balce-mo ' thy house ' from balay ' house '. In Ibanag the diph-
thong ay is pronounced ei, i. e., the first element is pronounced
e and not a, e. g.,patay 'dead,' ammay 'rice;' in Pangasinan
the first element of both the diphthongs ai and au seems to be
sometimes pronounced as e, as is indicated by the spelling, e. g.,
baley 'village, house,' ageo 'day.'

Generally speaking, cognate words have the same vowels in
all the different languages, a corresponding to a, i to i, o, u to
o, u. In a number of words, however, the vowels vary from
one language to another, in general, Tagalog i being repre-
sented in Bisayan and Bikol by o or u; in Ilokan, Pangasinan,
and Magindanao by e; in Ibanag and Pampangan by a; e. g.,

(passive

' rice '

' hear '

' room '

suffix)

'six'

Tag.

bigas

dingig

silid

-in

anim

Bis.

bugas

dungug

sulud

-on

unum

Bik.

dangog

-on

anum

Ilok.

bagas

dengeg

silid

-en

innem

Pang.

dengel

-en

anem

Mag.

begas

-en

anem

Iban.

baggat

-an

annam

Pamp.

-an

anam

1 In this article, unless otherwise stated, o and u, and i and e, are
respectively considered as equivalent.

332 F. It. Blake, [1906.

It is quite possible that this varied vocalism is the represen-
tation of a fourth primitive Philippine vowel, an indistinct
vowel like the Indo-European shewa,1 which in a similar way-
is represented by several different vowels in the various Indo-
European languages.2

In Sulu there is a vowel corresponding to the German il, e. g.,
tud 'very,' bukun 'it is not.' This, however, seems to be a
secondary Sulu development, words containing it having in some
cases by-forms with a or o, u, or corresponding to forms with
these vowels in the cognate languages, e. g. , maisug, maisog
' bold, intrepid ' (Bis. maisog) ; dakttp, dakap ' embrace ' (Malay
dakap).

Remarks on the Consonants.
Labials.

A final p in Ibanag regularly becomes the glottal catch (cf.
below, p. 336).

In some of the languages of the Northern Group p and b pass
at times into the spirants /and v. In Ibanag this change regu-
larly takes place when original p is followed by u or o, e. g.,
fulu ' ten ' (Pang, and Bis. polo, Magin. pulu, etc.). In Nabaloi
Igorot p and /"are often used interchangeably in the same word,
e. g. , apil or afil ' different ;' in others the p or f respectively
are constant, e. g.,palit 'dear,' andujit 'soft.'

In Bontoc Igorot, and also in the northern part of the Nabaloi
territory, there is a similar interchange between b and /, e. g.,
Bont. babayi andfafayi ' woman,' Nab. balei andfalei ' house.'

In Ibanag, just as pu becomes fu, bu becomes vu, the b, how-
ever, being still retained in spelling, e. g., buaya 'cayman.'

The sound v occurs in the one word asivi ' small ' in Sulu.

In Magindanao a b sometimes becomes w (written u), e. go
uato 'stone,' uata 'boy,' ualay 'house' (Tag. bato, bata, Bis.
balay); but it is sometimes preserved, e. g., babay 'woman,'
ngibu ' thousand' (Tag. babayi, libo).

In Nabaloi Igorot original wa regularly becomes gwa (written
gua or god), e. g., gualo 'eight,' asagoa 'spouse' (Tag. walo,
asawd).

1 Cf. Brandstetter, Tag. u. Mad., p. 34.

8 Cf. Brugmann, Qrundriss der Vergl. Gram, derindogerm. Sprachen,
2" Bearb., Strassburg, 1897, Bd. 1, p. 170.

Vol. xxvii.] Contributions to Philippine Grammar. 333

Dentals.

In Ibanag the combination ti regularly becomes si, e. g.,
sinakao ' robbed ' from a root takao with infix in. An original
final t regularly becomes the glottal catch (cf. below, p. 336).

A d of the other languages is regularly represented by the
affricative ch in Nabaloi Igorot, e. g., chalan 'road,' achalem
'deep' (Ilok. dalan, adalem], and in Bontoc Igorot the two
sounds are used interchangeably, as in the place names Chakong
or Dakong, Pudpudchog or Pudpuddog.

In Ibanag the combination di when not preceded by a (in
which case d > r, cf. below, becomes zi (written ji], e. g.,
jinakay ' leprous ' from the root dakay.

In Ibanag an s not followed by i seems to be regularly changed
to £, e. g., ataua 'spouse' (Tag. asawa), takit 'sickness' (Tag.
sakit) ta 'to, in the, etc.' (Tag. so) ; but si the personal article
(Tag. si).

One of the most complex chapters of Philippine phonology is
that which is concerned with the interrelations of the sounds d,
r,l.

In a number of words an Ibanag, Ilokan, Bikol, and Samaro-
Leytean r corresponds to a Tagalog, Bisayan (except Sam.-
Ley.), Pampangan Pangasinan, Magindanao and Sulu /, e. g.,

' write ' • ' silver ' ' thousand ' ' how much ?'

Iban. surat pirak ribu

Ilok. surat pirak ribo

Bik. surat pirak ribo pira

Sam. -Ley. surat ribo ('million') pira

Tag. sulat pilak libo

Bis. sulat pilak libo pila

Pamp. sulat libo pilan

Pang. sulat libo

Mag. sulat . pila

Sulu sulat pelak pela

As an original I is in general preserved without change in
most of the languages, r is here probably to be regarded as the
more original sound.

An original intervocalic I is lost in Tagalog and Sulu ; in Taga-
log it is represented by the glottal catch between the two
vowels, or a secondary h or semi-vowel is developed between

334 F. E. Blake, [1906.

them, e. g., daan 'road,' bahay ' house, ' powo 'ten' (Bis. dalan,
balay, polo); in Sulu the two vowels are contracted, e. g., dan
' road,' bai ' house,' poll ' ten.'

In Batan such an I seems regularly to become g, e. g., ago
'head,'^?0<7O 'ten,' bagay 'house,' bugan 'month' (Bis. olo,
polo, balay, bulan). The same change is perhaps illustrated in
Ibanag, Pangasinan, and Nabaloi piga 'how much?' (Bis. pila).
This change from I to g probably passed through the interme-
diate stages, guttural r, and the guttural sonant spirant like g
in the North German pronunciation of tage ' days.' The writing
g, indeed, may be simply an imperfect attempt to indicate the
latter sound.

In Nabaloi Igorot an I or r of the other languages is often
represented by d, e. g., idoko 'Ilokan' (Ilok. iloko], mabadin
'possible' (Ilok. inabalin), sudat 'write' (Ilok. surat, Tag.
sulat} .

In Batan the change from I to d is also found, e. g., dima
'five,' tatdo 'three' (Bis. lima, tatlo).

An intervocalic d is in many of the languages often changed
to r. So in Tagalog, Bisayan, Pampangan, Pangasinan, Ibanag,
Nabaloi, Batan, and Sulu; apparently not in Bikol, Ilokan and
Magindanao, cf.

Tag. narito ' is here ' from dito ' here. '
Bis. aruna, duna 'have.'
Pamp. kareni 'to these' from deni ' these.'
Pang, maronong ' wise ' from root donong.
Iban. ikarua ' second ' from root dua ' two.'
Nab. marikit 'pretty,' cf. Tag. dikit 'beauty.'
Bat. icharua 'second,' cf. Iban. dua 'two.'
Sulu ha ran ' on the road ' from dan ' road. '
Ilok. ida ' they ' from da.

Palatals and Gutturals.

In Ibanag final k regularly becomes the glottal catch (cf.
below, p. 336).

In Nabaloi Igorot k is often changed to the corresponding
surd spirant like GernAan ch, which may be represented by x,
e. g., ixamen 'mat' (Ilok. ikameri), asixen 'old man' (Pang.
asikeri) .

In Bontoc Igorot k and g are constantly interchanged, e. g.,
kulid or guild ' itch,' yeka or yega ' earthquake.'

Vol. xxvii.] Contributions to Philippine Grammar. 335

In Batan a & in juxtaposition to an i sound either before or
after it regularly becomes cA, e. g., icha- prefix which forms
ordinals, machi-. verbal prefix indicating accompaniment (Tag.
ika^ tnaki).

A g of the other languages is sometimes represented in Ilokan
by r, the change in all probability being the reverse of the pro-
cess illustrated by the change from I to g (cf. above, p. 334),
e. g., baro 'new,' rabiy 'night,' darat 'sand' (Tag. bago,
gab-i, dagat ' sea ') .

In Nabaloi Igorot y becomes regularly dy, e. g., dyo 'your,'
adyab 'call' (Ilok. yo, ayab).

Laryngeals.

Under this category the Philippine languages seem originally to
have possessed the glottal catch or laryngeal stop, and perhaps the
surd spirant h. The glottal catch is not recognized as a sepa-
rate sound in any of the works on the Philippine languages, but
its presence in many languages is indicated by the statements
of the grammars, and it probably occurs in all. For example,
the Tagalog grammars speak of final vowels with guttural
accent, of v-owels pronounced separately from the single conso-
nant which precedes them, and state that two juxtaposed vowels
stand in different syllables. These statements are to be under-
stood as follows. The hiatus between the two vowels in such
a word as doon ' there ' is of course identical with the glottal
catch : the fact that a vowel following a single consonant begins
a new syllable, as for example in gab-i 'night,' indicates that
between the two stands the glottal catch, which really begins
the second syllable : the so-called guttural accent of a final vowel
seems to be really a glottal catch after the vowel; when the
final vowel has at the same time the stress accent, the vowel is
mai'ked with a circumflex, e. g., wald 'is, has not;' when it is
unstressed, with a grave accent, e. g., wikd 'word, language.'
The glottal catch in Ibanag resulting from a final stop is of
course secondary (cf. below, p. 336).

The spirant h does not occur in Ibanag,1 Ilokan, Pangasinan,
Nabaloi, and Pampangan, but it may have been an original
Philippine sound, as it is preserved in the other Philippine
languages; cf.

1 Not given in the list of consonants.

336 F. M. Slake, [1906.

' wind ' ' dear ' ' hair ' ' before ' ' king'

Ilok. angin book ari

Pang. arap ari

Nab. buek

Pamp. angin mal

Tag. hangin mahal buhok harap hari

In Sulu a word with vocalic initial sometimes takes a secondary
A before it, e. g., hinom 'drink' (general Philippine inum).
In Tagalog, Bisayan, and Bikol an h is sometimes inserted
between a final vowel and the suffixes an and in (cf. below,
p. 337).

Other Phonetic Changes.

The final stops k, />, £, in Ibanag are, according to the state-
ment in the principal Ibanag grammar, not pronounced, but the
preceding vowel receives a peculiar aspiration.1 This clearly
points to the fact that these consonants have become the glottal
catch, like the final k in Malay.2 This pronunciation is indi-
cated in the grammar by placing a dash between the final con-
sonant and the preceding word, e. g., a-k ' I,' taki-t ' sickness.'
In the present article, however, the dash will be omitted for the
sake of convenience. Before this glottal catch the vowel a is
pronounced as o, though a is usually written (but cf. so-k 'I').

A characteristic feature of Ibanag and Ilokan is the doubling
of originally single consonants, e. g., Iban. battu 'stone,' annam
'six' (Tag. bato, anini), Ilok. adda 'be' (Malay add).

In Ibanag the final consonants £>, o?, </, n are assimilated to an
initial consonant in a following word, e. g. , kanak ku ' my food '
< kanan ku\ and all final consonants are pronounced in the
same syllable with an initial vowel of a following word, the
change from final surd stop to glottal catch not being made in

1 ' . . . quedan absolutamente sin sonido, y mudas; mas debe darse en
la vocal, que les precede, un golpe de aspiracion, que solo la voz de maes-
tro puede espresar, y ensenar debidamente.' Cf. De Cuevas, op. tit.,
p. 10.

2 Cf. Seidel, Prak. Gram. d. Mal. Spraehe, p. 5. It is interesting to
note that this final glottal catch is written ^j) (e. g., (Jpt ana' 'son')
which in Arabic in general is equal to k + glottal catch, and which in
some modern Arabic dialects has become simply the glottal catch. Cf.
Favre, op. cit., p. 12 ; Marsden, op. cit., p. 11 f. ; Wahrmund, Praktische
Gram, der neu-arab. Spraehe, Giessen, 1861, p. 11.

Vol. xxvii.] Contributions to Philippine Grammar. 337

this case, e. g., mcdugak 'I want' < mcelug ak, ayata mapia
* great joy ' < ayat a mapia.

In Tagalog after a final vowel the a of the particle ay and
the conjunction at may be elided, e. g., ang tawo'y mabuti
( < tawo ay] ' the man is good, ' ama't ina ( < ama at] ' father
and mother.' This elision may also take place after a final n,
the n being lost at the same time, e. g., iya'y mabuti (< iyan
ay] ' this is good,' ama ft ali (amain at} ' uncle and aunt.'

Many contractions and elisions take place in Pampangan, but
the statements of the grammar on this subject are very meagre
and unsatisfactory; cf., however, kana kita for kana ku ita,
iyeni for iya ini, meng for me ing, yang for ya ing^ totita for
toto ita.

In Tagalog the suffixes an and in are added directly to a
word ending in a final vowel with glottal catch, e. g., turo-an
'be taught;' after other final vowels an h is inserted, e. g.,
sabi-h-in 'be said,' una-h-an 'front part.' The same rule as
regards the insertion of the h probably applies also to Bisayan
and Bikol.

PRONOUNS DERIVED FROM PARTICLES.

The Philippine pronouns with regard to their origin may be
divided into two classes, viz.,

a) pronouns derived from monosyllabic particles,

b) pronouns derived from dissyllabic roots.

To the second class belong the indefinite pronouns, except
those derived from the interrogatives, e. g., T&g..&auntt 'a
little ' from the root unti, marami ' much ' from the root dami-
and the numerals, e. g., Tag. apat 'four,' lima ' five.'

To the first class belong .

a) the ligatures,

b) the articles,

c) the demonstrative pronouns,

d) the interrogative pronouns,

e) the personal pronouns.

The present investigation will be confined to the pronouns of
the first class, which will be treated in the order named.

Pronouns derived from particles consist either of a simple
root particle, or of a root particle combined with other prefixed
VOL. xxvii. 23

338 V. R. Blake, [1906.

and suffixed particles. The prefixes are usually articular in
character, and are employed to express case; the suffixes are
ordinarily derived from ligatures which stood originally between
the pronoun and the following word. Those pronouns that are
inflected distinguish in general three cases (cf. above, p. 324),
though sometimes two cases have identical forms.1

The Ligatures.

The ligatures that occur in the various languages are the fol-
lowing, viz. :

Tag. .

na -ng (-n) ay -y

Bis. (in gen.)

nga -ng -y

Bis. (Sam. -Ley.)

nga (-ng) -n

Bik.

na -ng

Pamp.

-ng -n a

Pang.

-n a ya -y

Ilok.

nga -n a

Iban.

nga a

Bat.

a

Ig. (Nab.)

a

Ig. (Bon.)

-n ay

Mag.

a i

Sulu

vy'

Bag.

i3

1 In the paradigms of the pronouns all those forms that are not given
as nominative or genitive in the grammars are grouped under the oblique.
As the oblique is often used as the equivalent of the genitive, e. g., Tag.
ama nong batd=sa bata-ng anna 'father of the boy,' forms that are
really oblique are often given by the grammars under the genitive, but
this will occasion little difficulty from a morphological point of view.
Sometimes the oblique forms are not given by the grammars, in which
case they are probably to be formed by prefixing the oblique of the defi-
nite article to the nominative.

2 Not specifically mentioned as" ligature by the grammars, though
examples of both occur.

3 Ka is also said to be a relative in the sentence dini doton kagi ka diri
nonga olitan 'here are words that it is not proper to reveal;' but it is
here in all probability simply the genitive of the definite article used
before the following clause, which modifies kagi 'words' just like a noun
in the genitive. .

Vol. xxvii.] Contributions to Philippine Grammar. 339

The forms with dash are used only after vowels, e. g., Bis.
taico-ng maayo ' good man,' the others, in general, after both
vowels and consonants, e. g., Bis. tawo nga maayo 'good
man,' maalam nga magtotoon 'learned teacher.' Tagalog «y,
-y are used only to join together two elements that stand to each
other in the relation of subject and predicate, and then only
when the subject precedes, e. g., ang tawo^y mabuti ' the man is
good.' Bisayan -y is also sometimes employed in this case, but
is also used as the equivalent of the other ligatures.1

The difference between -ng and -n in those languages which
possess both is difficult to determine. Often they seem to be
used interchangeably; in Pampangan -n is used especially before
an indefinite noun, like the signs of the indefinite object in Bis.,
and Iban. (cf. below, p. 345); e. g., huma ka-n danum. 'take
some water.' The choice of a and nga in those languages which
possess both seems to be regulated by euphonic considerations:
in Ilokan a must be used when the preceding word ends and the
following begins with a consonant, e. g., toy a balay 'this
bouse;' nga must.be used when the preceding word ends and
the following begins with a vowel a, e. g., dayta nga aso 'that
dog,' otherwise the two are used indiscriminately. Ilokan -n is
used before the adverbs sa and to, e. g., adda-n-sa 'is there
•perhaps,' adda-n-to 'will there be,' and has also various other
uses.2 In Pangasinan ya is used principally to join clauses, e. g.,
alam so libro ya wala-ds silid ' bring the book which is in the
room ;' -y is used as the equivalent of -n, and also as a substi-
tute for the nominative and genitive of the articles, e. g., talo-
ra-y silla=talora-n silla 'three chairs,' onla dia-y Antonio =
onla dia si Antonio ' come here Antonio.' Otherwise the four
forms are practically equivalent except that -n and -y are only

1 Cf. my paper, Differences between Tagalog and Bisayan, JAOS. xxv,
1904, p. 167 f.

2 Cf. Williams, Qrammatische Skizze der Ilokano-Sprache (disserta-
tion), Miinchen, 1904, p. 64 f.

3 Contracted from wala ed. In Pangasinan wala means ' to be,.' in
Tagalog and Bisayan it means ' not to be,' one being affirmative, the
other negative. A similar difference in meaning is presented in Semitic
by Hebrew j"f2{$ ' be willing,' Arabic ,j 'aba, Ethiopic fift? ; 'abafa

' be unwilling, refuse.'

340 F. R. Blake, [1906.

used after vowels. In the meager specimens of Bontoc Igorot
which are accessible -n is used after a vowel, ay after either
vowel or consonant, e. g., chuwa-n lasot 'two hundred,' chuwa
ay lifo 'two thousand,' slam ay poo 'ninety' (nine tens). In
Magindanao a is the usual ligature, i being used mainly after
interrogative words instead of the article *w, e. g., tingin i
midtalu salka 'who spoke to you?' In Sulu the ligature i
occurs sporadically, e. g., pela i bulan ' how many months ?' In
Bagobo the ligature is used as relative.

All these ligatures seem to be derived from the four particles
na, nga, a, and i. The forms -n and -ng are shortened respec-
tively from na and nga / i after a vowel forms the second ele-
ment of a diphthong and is then often written -y • ay and ya
are probably combinations of the two particles a and i- na and
nga are perhaps simply two forms of the same particle.

The Articles.

The Philippine languages possess not only a definite and
indefinite article, which are in the main equivalent to the corre-
sponding English articles, but also a personal and an inclusive
article (cf. above, p. 325).

Definite Article.

The forms of the definite ai'ticle in the various languages are,
viz. :

nom.

gen.

obi.

Tag. ang

nang

sa

Bis. (Ceb.) ang

sa

sa

Bis. (Hil.) ang (ing)

sang

sa

Bis. (Sam. -Ley.) an (in)

san, nan,

kanan sa

A

t

N

Bis. (Har.) ya, nan

sa, kan, et

Bik. an, si

nin, ninsi

, sa, kan

Pamp. sg. ing

ning

king

pi. ding

(ding)'

karing

Pang. sg. so, say

na

ed

1 Doubtless the correct form (cf. inc. art.), though not given by the
grammar.

Vol. xxvii.] Contributions to Philippine Grammar. 341

Pang. pi. | sara^' ira^' day ed saraF

( ray; so saray na saray

Ilok. sg. iti, ti ti iti

pi. dagiti dagiti kadagiti

Igor. (Nab.) e1 ne sun, chi

Iban. i na ta

Bat. u, su nu du, (u, su)2

gu j na, nu,3 sa sa, kana,
I kana, kanu3 kanu3

Sulu in sin, kan ha, pa

Bag. Janj i ka, ta ka, ta

Nominative.

Forms that are apparently root particles are ya, si, i, u, ti.
Ya is probably identical with the ligature ya; si is the personal
article used as definite (cf . below, p. 346) ; i is probably iden-
tical with the ligature i; ti is probably identical with the demon-
strative root particle ti (cf. below, p. 353).

Pangasinan, Batan, and Magindanao su, so, may be a root
particle, or u with an inflectional s borrowed from si, the nomi-
native of the personal article. Nabaloi e may be a variant form
of *, or a contraction of *ay identical with the ligature ay.

The forms ang, an, ing, in, yan seem to consist of the root
particles i and ya just described, and a which is probably iden-
tical with the ligature a, followed by the ligature -ng, -n,* which
is regularly used in these languages between two words that stand
to one another in the relation of modified and modifier, when the
first of the group ends in a vowel. Originally the root particle
was the article, and -ng, -n simply a connective, the -ng in Taga-

1 Also given as objective along with sun and chi.

* The oblique case forms are given as du, lu, su, but this is almost cer-
tainly a mistake for du, 1. u, su, 1. being an abbreviation for ' or,' as
in thf nominative su 1. u. No particle which could be connected with
this apparent lu occurs in any of the languages, except, perhaps, in
Magindanao (cf. below, p. 372) and there it seems to be a foreign ele-
ment. As these forms are headed 'dat., ac., abl.' the forms u, su are
probably nominatives used as accusatives (cf. below, p. 380, ft. nt. 1).

3 As these forms are given by Juanmarti in the paradigm only in con-
nection with the following plural particle manga, the u may be a modi-
fication of a due to the labial TO.

4 Cf. Brandstetter, Tag. u. Mad., p. 78.

342 F. R. Blake, [1906.

log a-ng tawo ' the man ' being no more a part of the article
than the -ng in ito-ng tawo 'this man.' The ligature, however,
has become an integral part of the article, as is shown by the
pronominal use of the article before an oblique case, e. g., Tag.
ang sa tawo 'that of the man,' where the ligature would ordi-
narily not be employed.

Pangasinan say seems to be a combination of a root particle
sa and the ligature i. Whether this sa is ultimately identical
with the sa of the oblique case (cf. below) is uncertain.

Harayan nan is probably borrowed from the genitive, being
doubtless identical with the Samaro-Leytean genitive nan (cf.
below).

Ilokan iti seems to be a combination of the simpler form ti
with an articular prefix i doubtless identical with Ibanag nomi-
native i, and ultimately with the ligature i. This i is to be dis-
tinguished from the prepositional i which occurs in the identical
oblique case iti (cf. below).

The plural forms in Pangasinan and Ilokan contain a plural
element da, ra, which is identical with the root particles da, ra,
la of the third person plural (cf. below, p. 381 f.). In Pangasi-
nan ra is always followed by the ligature i, and it also may
take the prefixes 'i, sa, and so sa, i being identical with the i
of Ilokan nominative iti, sa and so with the sa and so of the
Pangasinan singular. In Ilokan the plural is made by prefixing
dag, consisting of da-\-au additional pluralizing element g,1 to
the singular iti. In Pampangan ding, which like the singular
forms ang, ing, contains the ligature -ng, di may be a modifica-
tion of da due to the influence of the i of the singular, or it may
be an independent root particle (cf. below, p. 348).

Genitive.

The genitive forms that are apparently root particles are sa,
et, na, ti, ka. Ti is the same as the ti in the nominative ; sa, et
and ka belong rather to the oblique (cf . below) ; na is probably
identical with the ligature na.

Batan nu and perhaps Magindanao nu (cf., however, p. 341,
ft. nt.) may be a root particle, or it may be u with an inflec-

1 Cf . my article, The Bisayan Dialects, cited above, p. 127.

Vol. xxvii.] Contributions to Philippine Grammar. 343

tional n borrowed from na. Nabaloi ne is either a variant form
of ni, or na -f- the ligature i. Bagobo ta has apparently no rep-
resentative in any of the other languages, Ibanag ta (obi.) rep-
resenting an original sa.

The forms nang, nan, sang, san, consist of the root
particles na and sa followed by the ligature -na, -n. The
forms ning, nin consist of a similar combination of the ligature
with a particle ni, which may itself be a root particle, or the
root particle i with an inflectional n borrowed from na.

In Bikol ninsi the genitive is prefixed to the nominative.
Sulu sin is probably the nominative in with an inflectional s
borrowed from the genitive and oblique sa (cf. below, p. 354).
The forms kan, kanan, kana, kanu, seem to be borrowed from
the oblique (cf. below).

The plural forms in Pampangan and Ilokan are identical with
the nominative. In Pangasinan the form day (= plural particle
da -\- ligature i} is probably more original than ray of the nom-
inative, where the r is doubtless due to the analogy of the
other nominative forms (cf. below, p. 382): in the form .na.
saray, na, the genitive singular, is prefixed as genitive case sign
to the nominative.

Oblique.

The forms that are apparently root particles are sa (including
Iban. ta, cf. above, p. 333), et, ed, *di (Nab. chi, cf. above,
p. 333), ha, pa, ka. Et and ed are probably identical, and are
perhaps to be connected with di; di is doubtless identical with
the Malay preposition di ' in ;' Sulu ha is perhaps to be connected
with the Sulu active verbal suffixes a, ha just as the i of Ilokan
i-ti is probably identical with the active verbal suffix * of Malay
(cf . below) ; Sulu pa, which indicates motion towards, is per-
haps identical with the verbal prefix pa, which is used to form
verbs of motion, e. g., Tag. pa-rito 'come here' from dito
'here,' pa-sa-Maynila 'go to Manila' from sa-Maynila 'in
Manila ;' ka is. doubtless identical with the Malay preposition ka
' to, towards.' For Bagobo ta cf. under genitive.

The form kan consists of the root particle ka-\- ligature n;
Pampangan king consists of the ligature ng and a root par-
ticle ki which is probably modified from ka under the influence

344 . F. H. Blake, [1906.

of the i vocalism of the other forms; Bataii du is probably u
with an inflectional d borrowed from the particle di; Nabaloi
sun consists perhaps of the root particle u with an inflectional s
as in Sulu genitive sin, and ligature n; Ilokan iti consists of the
root particle ti, and a prepositional il identical with the i of the
Tagalog adverbs and prepositions sa i-babao ' over,' sa i-taas
'above,' sa i-babd 'below,' sa i-lalim 'underneath,' which con-
sist of a root preceded by two prepositional elements i and sa.
This * may be identical with the prefix i of the passive, just as
in Malay di is both preposition and passive prefix.2 The same
preposition i seems also to occur as active suffix in Malay, where
it is equivalent to the active suffix kan, connected with the prep-
osition ka.3

In Samaro-Leytean kanan, Magindanao kana, kanu, ka is
prefixed as oblique case sign to the genitive; in the plural of
Pampangan and Ilokan to the nominative-genitive. In the
Pagnasinan plural, ed is prefixed as oblique case sign to the
nominative.

For the Batan forms u and su, which are identical with the
nominative, cf. above, p. 341, ft. nt. 2.

Indefinite Article.

The idea of the indefinite article, whenever it is not indicated
sufficiently by the nature of the construction, is expressed by
the numeral one, which unlike the numerals from ' two ' upward
seems to be derived from pronominal particles. This numeral,
however, is not inflected, but expresses its case by means of an
inflected word placed before it. Its forms in the various lan-
guages are, viz. :

Tag. isa Ilok. meysa

Bis. (Ceb.) usa Igor. (Nab.) saxei

Bis. (Hil.) isa, usa Igor. (Bon.) isa

Bis. (Sam. -Ley.) usa Iban. itte, tadday

Bik. saro Mag. isa

Pamp. isa, metong Sulu jsa

Pang. isa, sakey Bag. sabbad

1 Cf. Brandstetter, Tag. u. Mad., p. 78.

2 Cf. Favre, Grammaire de la langue malaise, Vienne, 1876, pp. 132,
158. Seidel, op. cit., pp. 52, 66.

3 Cf. Marsden, op. cit., p. 56.

Vol. xxvii.] Contributions to Philippine Grammar. 345

All of these forms except Pampangan metong, which is not
clear, are derived from a root particle sa, which is used for
* one ' in Javanese and Malay. The initial i of isa is probably
the same articular i which has already been seen in Ilokan
nominative iti, and which has an extensive use in the demon-
strative and personal pronouns (cf. below).

Ibanag itte probably consists of *itta -f- the ligature i, just as
due ' two ' is probably equivalent to dua-\-i; *itta is doubtless
derived from *ita (=isa, cf. above, p. 333), with the doubling
which is so characteristic of Ibanag (cf. above, p. 336). In
Ilokan instead of the simple i, mey, contracted from mai, a
combination of i and the adjectival prefix ma, is used, just as
it is employed instead of the simple i of many of the other lan-
guages, in the formation of the ordinals, e. g., Ilok. maikapat
'fourth' (Tag. ikapat).

The u of Bisayan usa is probably identical with Batan u (cf .
also below, pp. 354, 361).

Ibanag tadday (<.*sadday, cf. above, p. 333) is perhaps to
be analyzed as root particle sa -j- da, root particle of third per-
son plural, -f- ligature i, the doubling being phonetic. The
second elements of Bikol sa-ro, Pangasinan sa-key, Nabaloi sa-
xei, Bagobo sa-bbad are, perhaps, similar in nature to the
numeral coefficients of Malay.1

Under the head of indefinite articles are best considered those
particles which indicate the indefinite object of an action. In
Cebuan these are ug, more rarely ok; in Hiliguayna and Samaro-
Leytean sing and sin respectively: e. g., Cebuan kumuha ka
ug tubig 'take some water.' The two last are probably iden-
tical in formation with sin, the genitive of the definite article in
Sulu. In Ibanag a particle tu, perhaps identical with Batan and
Magindanao su, Pangasinan so, is employed, e. g., apam mu tu
kanak ku 'bring me something to eat.'

Personal Article.

The forms of the personal article in the various languages
are, viz. :

1 Cf. Favre, op. cit., p. 71 f. ; Maxwell, Manual of the Malay Lang.,
London, 1902. p. 70 f.

346

F. E. Slake,

[1906.

nom

. gen.

obi.

Tag.

si

ni

kay

Bis.

si

jni

j kay (Hil., Har.)

(•kanan (Sam. -Ley.)

( kan (Ceb., Sam.-

Ley., Hil.)

Bik.

si

ni

ki

Pamp.

i

nan

kan

Pang.

si

nen

ed

Ilok.

si,

ni ni

ken

Igor. (Nab.)

si1

nan, ne

sun, sikan PV%

Iban.

si

ni

ta, takkua, tlfk-

kuani

Bat.

si

ni

di

Mag.

si

ni, kani

kani

Sulu

hi

Bag.

si

ni

kan

The usual nominative is the root particle si. In Pampangan
i, identical with the root particle i of the definite article, is
employed, and with this, Sulu hi is perhaps identical, as a second-
ary h is sometimes developed before an initial vowel (cf. above,
p. 336). Ilokan ni is derived from the genitive (cf. below,
p. 387).

The usual genitive is the particle ni (cf. above, p. 343). The
forms nan, kanan, ne are to be explained like nan, kanan, ne
of the definite ai'ticle. Pangasinan nen seems to be the articu-
lar root particle na-\-ihe ligatures i and n, *nain being contracted
to nen (cf. above, p. 331). Magihdanao kani belongs rather to
the oblique.

In the oblique case most of the forms are based on the articu-
lar root particle ka : kan consists of ka -(- the ligature n ; kay,
of ka -\- the ligature y ; Ilokan ken is probably to be analyzed,
like Pangasinan nen, as ka-{-i-\-n ; Magindanao kani consists
of ka -f- the genitive ni : Nabaloi sikan is kan with what is
apparently the nominative si prefixed. Bikol ki seems to be a
modification of ka due to the influence of the i in the other case
forms si and ni. Pangasinan does not distinguish between the

1 Also given as objective along with sun and sikan.

Vol. xxvii.] Contributions to Philippine Grammar. 347

definite and personal articles in the oblique case, ed serving for
both, and in Ibanag the oblique case ta of the definite article is
used either alone or in the compound forms takkua, takkuani
as the oblique of the personal article.- The kua in the com-
pounds just mentioned is a noun meaning 'thing, possession,'
which is used in Ilokan and Ibanag in conjunction with the post-
positive genitives of the personal pronouns to form possessive
pronouns, e. g., kua-k ' mine,' kua-m 'thine,' etc.1; ni is the
genitive of the personal article. Batan di is doubtless identical
i chi, Malay di (cf. above, p. 343).

Inclusive Article.

The forms of the inclusive article, in so far as they are given
by the various grammars, are as follows, viz. :

nom. gen. obi.

Tag. sina nina kana

Bis. (Ceb.) sa na ka

Bis. (Hil.) sanday nanday kanday

sila nay ila nay sa ila nay

sila ni ila ni sa ila ni

sila si nila ni kanila ni
sila sa2

Bis. (Sam. -Ley.) sira nira kanda
TJ.I j sa, sainda sa, na, ninda na na, sainda na

( sinda sa

Pamp. di di kari

( sara di ( da di -,.

Pang. ed sara di

( sikara di ( na sara di

Ilok. da3 da3

Iban. da da takkuara

The Tagalog forms are compounded of a particle na, perhaps
identical with the ligature na, and the case forms of the per-
sonal article, ka being used instead of ka-y.

1 Cf. Naves, op. cit., p. 85; De Cuevas, op. cit., p. 77.

2 Cf . Mentrida and Aparicio, op. cit. , p. 10.

3 Cf. Naves, op. cit., pp, 80, 121.

348 F. R. Blake, . [1906.

Cebuan and Bikol sa, which occurs also in Hiliguayna silo, sa,
seems to be a modification of si, a perhaps being borrowed from
the plural particle da, where it was felt as a plural ending.
The other case forms na and ka were then made on the basis of
sa, the nominative case sign s being changed respectively to n
and k.

In many instances the inclusive article is based on the pro-
noun of the third person plural (cf. below, p. 379 ff.). In Hili-
guayna, sanday is the pronoun sanda ' they ' + the ligature i;
nanday and kanday are made on this basis by simply engaging
the case sign : sila nay and its declined forms is a combination
of the pronoun sila ' they ' with a particle nay, probably na,
identical with na in Tagalog si-na, -\- the ligature i : for sila sa
cf . below. The other Hiligu'ayna forms consist of sila followed
by the nominative or genitive of the personal article.

Samaro-Leytean sira, nira are identical with the correspond-
ing forms of the pronoun of the third person plural; the oblique
kanda is to be connected with the stem of Hiliguayna sanday.
The Bikol forms like sinda sa, ninda na, etc., and Hiliguayna
sila sa, consist of the forms of the pronoun of the third person
plural followed by the nominative or genitive of the inclusive
article sa ; sainda sa, if it is not a typographical error, owes its
initial syllable sa to the influence of the inclusive article sa. Ilo-
kan and Ibanag da is identical with the root particle da of the pro-
noun of the third person plural. The sign of the oblique case in
Ibanag, viz., takkua- is to be explained as in the case of the per-
sonal article. Pampangan di may be a modification of da due to
the influence of the prevailing i vocalism of the articular forms,
but the occurrence of di as the final element of the inclusive
article in Pangasinan, where there is no such influence, seems
to indicate that di is an independent particle: kari is di
with prefixed case sign ka. In Pangasinan the inclusive
article is made by prefixing to di, forms derived from the pro-
noun of the third person plural or from the plural of the defi-
nite article : sikara is identical with one of the pronouns of the
third person plural; the other elements prefixed to di are the
corresponding case forms of the plural of the definite article
without the ligature -y.

Vol. xxvii.] Contributions to Philippine Grammar. 349

Demonstrative Pronouns.

With regard to their meaning, the demonstrative pronouns
may be divided into two general classes, A) the nearer demon-
stratives, which may be translated by ' this ;' B) the more remote
demonstratives, which may be translated by 'that.' Within these
two general groups, especially in the second, there are in most
of the languages further subdivisions of meaning. For example,
in Tagalog, yari indicates something that is nearer the speaker
than the person addressed, ito, something that is near both,
iyan, something that is nearer the person addressed than the
speaker, and yaon, something that is distant from both.

These more minute distinctions, however, are of practically
no importance from a morphological point of view, and even in
the case of the two main semantic groups, we often find what is
nearer demonstrative in one, used as remoter demonstrative in
another, e. g., Tagalog ito 'this,' Samaro-Leytean ito 'that.'

In general the demonstratives are made up of three elements,
viz., (a) demonstrative root particles; (b) prefixed pai'ticles,
usually of articular origin, which denote case and sometimes
number, and which will be known as case indicating particles;
(c) connective particles or ligatures, which are of two kinds,
those that connect case indicating prefix with the root particle,
and those that are suffixal in character, representing what was
originally ligature between the demonstrative and following
noun. Sometimes a demonstrative seems to contain two root
demonstrative particles, viz., in Magindanao en-tu and in the
Ibanag forms ending in -ye. In Bagobo an adverbial particle
go appears as final element in some of the demonstratives.

The forms of the demonstrative pronouns in the various lan-
guages are, viz. :

350

[1906.

PQ

03

saman
saman

aman,

03

—
03

00

O

„

O

03

0 5

u

00

• a
o

o

.b

'a

4-3

O
*J

•4J1
0
-4^

r3

4-3

O

-4-3

'a

a §

O -4J

x -a

o3

00

0 .3

a

03

a"

03

a

OJ

OQ

man; s<
tonman, n

Q
rt

E
3

OQ
TJ

OJ

saraman,

raman ;

C
98

1

a"

o

03

.b

0

4-3

o

O

-4-3

.3

a

c

a
a"

nd
o3

03
oo

0 c3

-4-3 O2

of

o3

•4-3

0

o

o

-4-3

4-2

03
oo

^ ™

o"

*

a"
o

08

a

11

^o"

—
^

03

O
•4-J

n3
o3

• rH

II

0 |

rC* c^

S ' fiP 5
jf tf"3 S

^ CQ

03 '"S * ?
n Q 43 b

=| ."S 03 O> o3

a ^

r*

a

i>J 00

c3 OQ

"•^3 ^ [^ ^

."S a f^ ^ r^

a

03

a

a"

a

03

—

o3

'oo

03

00

a

I

03

-4-3

03
D

a *

H

K' 03

a

03

-4-3

03

:~
03

00

0
oo

o3 S?

03

03

o a

• i^

a .2

CC

p

^

• *\

a

03

a a

o3 o3

a *^^

a

03

niana

""" 'oo

11

!>j oo

Ml

a" 03
rn ce a >*

1 ^2 o3 "2
**->

"a
a "_, "_, 03 ^*

C 03 '•H ^ O ^
n» t>^ 03 03 PH O3
K^ .I-H QP <1> o3 ^

^ C Jd T3 _b^ °^

.1-1 ^- 1— W !-»( „,;

a o3

03 -tf

-u a

0

s

B8

r.

saratan

a

03
g

5 .§ •§ * |

:s §

q o3

'^

^N ^w «—

"a

"oo

«8

o3 ^
• i—i 03
oo

of

98

>,

7i

5
•yj

0

^

[S

'a

'oo

"S

03

J3

of o3 'S

eg

03-

BE

'?»

r:

OQ •-<

03
00

:il

'a

rt

OB

• •-1 o3

'a .PH
•r-H <PH . < a

SOS "— .1-1
tp* rM .5 a

"a
'a

03

*a
'a 2

0» 03

03 o3
„ " ^

1 S 'a
o? 3

00

rt

oo

-=

-^

of
03

OQ

>.

71

zi

d d ^ 6 c5

^ O

^

C

^ O

^ O (z; O

O

O O

^ O

C

*

: . °'

d

q>

d

d

d ^

s

*

t

rg

^

a

m'

03

a,

Q

" x

3

ee

o .

PQ

fee

a

03

Vol. xxvii.] Contributions to Philippine Grammar.

351

z

1

~~

03

as*

a

So

•S

_

tf>

03

>J *"§

fl

5

CD

S o3 t ^^

03

—

danman; i
man

e -p3 •- .S

S ^ J43 £> '^

•~ £r- - ^ 'So

g .2 ^J .£ 03
r^ CD '— 03 o3

a o

i-H 43

a" CD CD ^

O r^ r^ "r~ » r^ *r~s ^ !H ^H 2 C
CO ^s ^i ^5 ^> ^^ ^j ^s ^^ «^^ ^a

S

o

co
OS

^3
0

"S
o

'cs

43

03
CD

a

~Y

1.

O

'53

•§

Tc

CO

S

o

CO

^3

3

i

'S

^"i

0

fi •

^.

43

"- «* .„,

c

0

.2

X!

2

CD

^

"""i

^

._r -5 ;g 'bo

T9 "" ,S c3

CD fl >% O

c: :P5 ^ S ^ s bo

bo

sT

i

CD

4J

CD

co
03

o3 r^ 03 o3

^=?"o343 ^C"S'3
t->-»t>j>^ o3a>.rt43

o
1.

in"

a

U

S

CD

a

oS

o3

•8

s
s

'5

0

43

03

g

oS
8

03

43
OQ
p

1

a

55

5

o

0

^

•3

CD

03

£~^

oS

r

—

f}

**

G

a

a

d

03

>9

I3' 2"

C3 „ ,_j _a -^

^ d T .^ J ^

50 ^'-'Sb'S

f"^ oS Q oS o3

43
CO

03 03 .-H ,
-i-a f- f-> •-! =i G !^»
03 o3 oS l* O CD O
cc ^j t>-, S o3 -i-i 43

itratives, viz

S'

•z

bC

5

co

'o

intervocali(

d
•a

5

HH

0

d

^

— . —

HU

a

'3

e

13

0

'£

£*•

a

H.

.Ss

'3

CD

C^

rG

*^

*

0

CD

^

^

O

_

z

0

pC

*

9

S

02

CD

<z

_c

03

^3

t^

—

co

0)

CO

-2

g

a

0

2

2

*^

•»•

c

c

j

1

^2

0

2

03

.s v Y

CD

• S

03

03

oi

co

X

03
ea

oS

a

03

'3

O

43

oS "^ o

'S !*"1 3 'So

r^ 03 0 OS "^

.^ <»

""^ 'S 'o '** S is S^ ^
o3 © *o ^ o3 C

cc i^ pj o3

'S03 SSScSS5*
*^l *N o3 oS 03 ^ ^^ K^
.2 S^sHCP^^Ojj
03 oSoScS^CS'ioSosS'^ *i "^

CC f^-j H 43 rg pj rQ 4j +j S fl

03

CO
CD

03

S

3

Cebuan paradigm
is can represents

4^

oo '**

CO C
. CD

o .S
"* TS

A

CO~

0

• o d o o
o

jz d o d

Jz; O

i

^z

=
.5

<s'l

^|

5^' ^'

fei ^

CD

=

^

e £

^ •&

^ u

"7- ^

o3

e

CD

J

,*3) O

so **"

CD q

00 «H

03

CD

>

-"

^ ^

-S ^

H

5

d

M

8 ^

- S

-g

S)l -'I "3 ^

•?

•a

PJ

OB ^ rA S ^r> 'A
HH HH W ^ CO W

352 F. R. Slake, [1906.

The root particles of the demonstrative are di, ni, to, ya, au,
na, ti, ta, ma, en, and possibly a (cf. below, under ya and na).
The particles ya, na, a are probably identical, respectively, with
the ligatures, ya, na, a.

Di, which when intervocalic appears in many languages as
ri, is found in Tag. ya-ri, yi-ri, ye-ri, ni-ri, all the forms of
Ceb. ka-ri, Iban. ya-ri, yu-ri, Bat. u-ri, Ilok. clay-di, de-di-
ay, and in Nab. sai-di-ai, sa-di-ai. The root particle is found
uncombined only in Ilokan. Ilokan dediay is daydi -f- the
double ligature ay, day being contracted to de. In Ibanag the
forms yad and yud seem to to shortened from yari and yuri
respectively; and these shortened forms were then augmented
by the addition of a, probably the ligature a, viz., yara, yura,
and ye, probably contracted from the ya which occurs as the
pronoun of the third person + the ligature -y, viz., yajje
(<*yad-ye), yujje (<*yud-ye).

Ni* occurs in Tagalog oblique di-ni and all the forms of
Cebuan k-i-ni, Hiliguayna, Samaro-Leytean, Bikol, Pampan-
gan, Sulu, and Bagobo i-ni, and Magindanao i-ni-a.

To occurs in all the forms of Tagalog i-to, Cebuan k-i-to-t,
Hiliguayna i-to-n, Samaro-Leytean i-to, Ilokan day-to-y,
Ibanag ya-tu-n, ya-tu-n-ye, in Bagobo to, to-y, to-i-go, probably
in Magindanao en-tu, Sulu ie-tu, and in all the forms derived
from the stems, Cebuan, Hiliguayna, and Samaro-Leytean ad-
to, Harayan ag-to, and Bikol id-to. The second t of the Cebuan
forms kitot, etc.; kadtot, etc., is probably due to incomplete
reduplication. The final ye of the Ibanag forms is to be explained
as the ye in yajje (<.*yad-ye}. Bagobo -go is an emphatic parti-
cle which is also used in sentences with non-verbal predicate,
e. g., si kona mapia-go manobo 'you are a good man.' The
en- of Magindanao entu seems to be the same as the en in Sulu
i-en:

Ya occurs in all the forms of Tagalog, Bikol, and Pampan-
gan i-ya-n ; of Pangasinan sa-ya; in the Ibanag final syllable
ye for *ya-i in yoye, yajje, yujje, yojje, yatunye; and perhaps
in Harayan dia (all forms), Nabaloi iai, Batan nooya, and
Bagobo yango. Harayan dia probably stands for *di-ya (cf.

1 Ny, the definite article in Madagascar!, is probably identical with
this root particle, cf . Parker, op. eit. , p. 45.

Vol. xxvii.] Contributions to Philippine Grammar. 353

sia—siya, p. 381, below). Nabaloi iaimay be simply a shortened
form of saidiai, sadiai, or it may be analyzed as ia -f- ligature
«', or as a root particle a preceded by articular i, and followed
by the ligature i. Bagobo yango contains the particle go and
an element yan which is probably ya -\- the ligature n.

Au occurs in Tagalog y-ao-n, ni-y-ao-n, and in all the forms
of Ibanag y-au, yoye (<^*y-au^ye), Batan au, and Sulu i-au-n.
In the Tagalog forms yoon, niyo.on, the first element of the a of
ao seems to have been assimilated to o; nion is contracted from
niyoon; noon and doon seemed to be formed on the basis of
yoon, the case signs n and d being substituted* for y. The
Ibanag forms yod, yojje (<*yod-ye), yora present a blend of
you and the forms with d, having the vocalism of yau, and the
d of the latter forms.

No, is found in all the forms of Cebuan ka-na, H^iliguayna
ya-na, Magindanao a-na-n, and perhaps of Harayan dan if it is
a shortened form of *da-na. In the Harayan forms, however,
the root particle may be a, identical with the root particle of
Tagalog ang.

Ti, identical with the ti of the Ilokan definite article, occurs
in all the forms of Pampangan i-ti.

Ta occurs in all the forms of Pampangan i-ta, Pangasinan
sa-ta-w, -Ilokan day-ta, and in Nabaloi sa-ta-n, su-ta-n and i-ta-n.

Ma occurs in all the forms of Pangasinan and Nabaloi sa-
ma-n, etc.

En occurs in Sulu i-en and Magindanao en-tu, the tu of the
latter form 'being the demonstrative root particle to.

The case indicating pi-efixes of the demonstratives are, in
Tagalog i, ni, n, di, d ; in Cebuan k, «a, s, di, d, i, kan, ni •
in Hiliguayna and Samaro-Leytean z, *, sa ; in Harayan d, c?e(?),
Ara, k • in Pampangan «, ni^ ka, da • in Pangasinan sa, a, i, to,
da, so, na, ed • in Ilokan i, da. dag / in Nabaloi sa, su, i ; in
Ibanag i, n, t, da, an (?), takka, takkua ; in Batan u; in
Magindanao a, i, in Sulu and Bagobo i.

The prefix i (y before a vowel) which is found in most, of the
languages as the sign of the nominative case is apparently iden-
tical with the articular root particle i.1 The * of the oblique

1 This i is found also in the Malay demonstratives i-ni ' this,' i-tu
'that;' and in all the demonstratives in Madagascan, e. g., i-ty 'this,
cf. Parker, op. cit., p. 41.
VOL. xxvii. 24

354 F. E. Blake, [1906.

singular in Ilokan, however, is probably a prepositional particle
(cf. above, p. 344). Batan u is identical with the u of the defi-
nite article; ni, the most common sign of the genitive, is identi-
cal with the ni of the personal article: sa, ka, kan, di and tak-
kua are identical with articular oblique forms sa, ka, kan, di and
takkua ; da is the root particle of the third person plural that
has already been met with in the articles ; in Ilokan the da seems
to have lost its plural force, being used as a simple initial demon-
strative element in both singular and plural, the plural being
indicated by da with an additional plural sign g, viz., dag;
Pangasinan so, na, ed are case forms of the definite article ; Pan-
gasinan and Nabaloi sa is identical with the sa of the article say;
Pangasinan and Magindanao a is probably the same a as in Tag-
alog ang ; Pangasinan to is the genitive of the pronoun of the
third person singular; Nabaloi su is to be connected with the
articles, Pangasinan so, Magindanao su ; Pampangan ke and de
are contracted respectively from ka-i and da-i, with articular i ;
Ibanag takka is a double oblique sign, standing for *ta-ka <^
*sa-ka (cf . above, pp. 333, 336) ; for an of the Ibanag nomi-
native plural, cf. below, p. 356 f.

When the case sign consists of a single consonant it is usually
to be explained as derived from forms with a fuller case sign
under the influence of proportional analogy. For example, the
s of the genitive-oblique in Hiliguayna and Samaro-Leytean is
probably derived from sa, a series like Samaro-Leytean adto,
sadto (<sa-adto) giving rise to forms like sini, sito on the basis
of ini, ito. Similarly Tagalog n and d in the forms noon, doon
are probably derived from the fuller forms ni and di ; Cebuan
and Harayan k, sa, s, and dof the nominative are probably based
on the articular oblique forms ka, sa and di (for the use of
these oblique forms in the nominative, cf . below, p. 388) ; Har-
ayan k in kagto is certainly derived from the oblique form ka;
Ibanag n and t of the genitive and oblique respectively seem
to be derived from the articular forms na and ta.

Sometimes the case signs are prefixed to a case form, some-
times directly to the demonstrative root particle, sometimes to
a demonstrative stem consisting of a root particle combined
with a prefix, usually articular in character. These prefixes are
a, i and u identical with articular a, i and u (cf. below, p. 3iil),
and ad, id, ag of uncertain origin. It is not impossible that

Vol. xxvii.] Contributions to Philippine Grammar. 355

ad, id and ag contain the articular particles a and i followed
by a connective g, which is preserved in Harayan, and which
is perhaps related to tig, the sign of the indefinite object in
Cebuan ; Cebuan and Bikol d representing a partial assimilation
of this g to the following dental t. Bikol id is probably modi-
fied from ad under the influence of the initial i of the other
demonstratives ini, iyan.

In Tagalog the case signs are usually prefixed directly to the
demonstrative root particle, e. g., i-to, ni-to, di-to. In the
nominative y-a-ri, y-i-ri, the case sign i is prefixed to the stems
a-ri, i-ri ; yeri seems to be a modification of yiri due to the r/1
in the genitives ni-yaon, ni-yoon the case sign is prefixed to the
nominative ; for other forms of yaon, cf . above.

In Cebuan the case signs are all prefixed to demonstrative
stems, viz., a-ri, i-ni, i-tot, ad-to, a-na, except in the nomina-
tives itot, diri^ where the case signs i, di are prefixed to the
root particle. The forms ari, adto are without case signs.

In Hiliguayna the nominative case sign is prefixed to the root
particle in i-ni, i-na and i-ton, but to a demonstrative stem in
y-ana, y-adto • the genitive-oblique sign s is prefixed to the
first three nominatives; in the others, it replaces the nominative
case sign ; another series of genitive oblique forms are made by
prefixing sa to these forms with s.

In Samaro-Leytean the nominative case sign is prefixed to the
root particle in i-ni, i-to, but to a demonstrative stem in y-adto ;
the form adto, as in Cebuan, is without case sign ; the genitive-
oblique case sign s is prefixed to the nominatives ini, ito, adto,
the genitive-oblique sign sa, as in Hiliguayna, to the forms
with s.

In Harayan the nominative is made by prefixing d to the
stems an (cf. above, p. 353) and ag-to, and di to the root parti-
cle in dia (cf. above, p. 352) ; the genitive-oblique is made by
prefixing the case sign to the nominative, except in kagto, where
the nominative case sign is simply changed to k.

In Bikol the nominative case sign is prefixed to the root
particle in i-ni and i-yan • the i in idto is probably not the
prefix i,' but is due to analogical influences (cf . above) ; the
genitive-oblique is made by prefixing the case sign ka to what
is apparently the nominative.

1 Cf . Nabaloi era ' they ' for *ira, p. 383.

356 F. ft. make, [1906.

In Pampangan the articular particles of the nominative and
genitive singular are prefixed directly to the stem, e. g., i-ni,
ni-ni / that of the oblique singular, to either the genitive or
nominative, e. g., ka-nini, keni (<*ka-ini): to form the nomi-
native-genitive plural the plural particle da is prefixed to the
nominative singular, e. g., deni (<*<?a-tm), and the oblique
plural is made from this form by prefixing ka, e. g., ka-reni.

In Pangasinan the nominative singular may begin with the
stem as in Samaro-Leytean adto, or may take the articular pre-
fixes sa, «, and i, e. g., tan, sa-tan, a-tan, i-tan ; the nominative
plural is formed by prefixing to the stem the particle da or its
combinations with the articular prefixes sa and i, viz., sara, ira,
e. g., ra-tan, sa-ra-tan, i-ra-tan: the genitive, singular and plu-
ral, is formed by prefixing to the stem the genitive singular and
plural, respectively of the pronoun of the third person, the two
elements being joined by the ligature n, e. g., to-n-tan, da-n-tan :
the oblique, singular and plural, is always, and the other cases
may be, indicated by the nominative with initial sa preceded by
the forms of the definite article, e. g., ed satan, ed saratan • so
satan, na satan, so saratan, na saratan.

In Ilokan the nominative-genitive singular may begin with
the root particle, or may take the prefix day, a combination of
the plural particle da and the articular prefix i, e. g., toy,
da-y-toy ; the nominative-genitive plural is formed from the
nominative singular by substituting for da a particle dag consist-
ing of plural particle da -j- a pluraliziug g ; the da of the sin-
gular is borrowed from the plural, where it has apparently lost
its plural significance, owing to the presence of an additional
plural sign g ; the oblique singular is made by prefixing a prepo-
sitional particle i (cf. above, p. 344) to the short form of the
nominative, e. g., i-toy, or by placing the oblique of the article
before the long form of the nominative, as in Pangasinan, e.g.,
iti daytoy ; the oblique plural is formed by prefixing ka to the
nominative-genitive plural, e. g., ka-dagitoy.

In Ibanag the case signs of the singular are prefixed directly
to the root particle as in y-au, or to a demonstrative stem as in
y-a-ri, y-u-ri • in the plural the case signs are prefixed to the
genitive singular, the sign of the nominative being the plural
particle da, or a particle an of uncertain identity, that of the
genitive, the particle da, that of the oblique one of the com-

Vol. xxvii.] Contributions to Philippine Grammar. 357

pounds takkara, takkuara, equivalent to the particle da
preceded by the oblique case sign takka, takkua. In the nomi-
native singular there are also a series of forms without the case
sign y, e. g., au: it is not impossible that the plural forms like
annau may have been made from plural forms \ikedanau by
dropping the initial d after the analogy of pairs like yau , au in
the singular; in this case the doubling of the n would be
phonetic.

The declension of the demonstratives in the remaining lan-
guages is not given in the grammars. The cases are probably
indicated by the forms of the definite article. The nominative
case signs a, i and u occur prefixed to the root particle in Nab-
aloi i-tan, i-man • Magindanao, a-nan, i-nia / Sulu and Bagobo
i-ni, Sulu i-en, i-aun • Batan u-ri: Sulu ietu is perhaps a mod-
ification of *i-tu with articular «', due to the influence of the
initial ie of the synonymous ien : for Nabaloi iai^ Bagobo yangot
Of. above, p. 352 f. The nominative case signs sa and su are
found in the remaining Nabaloi forms prefixed to the root
particle except in saidiai, which probably contains sa prefixed
to a stem beginning with articular i.

The connective particles which are used in the formation of
the demonstratives are n, i (y), a and ai (ay).

JVoccurs as final in all the forms of Tagalog, Bikol, and Pam-
pangan iya-n, Tagalog yao-n, Hiliguayna ito-n, Pangasinan
sata-n and sama-n, Ibanag yatu-n, in Nabaloi sata-n, suta-n,
ita-n, sama-n, ima-n, Magindanao ana-n, and Sulu iau-n: the
same n occurs followed by another element in all the forms of
Ibanag yatu-n-ye, and in Bagobo ya-n-go. This n is also used
in Pangasinan to connect the genitive case signs to and da with
the root particle (cf. above, p. 356), The n of all the forms of
Harayan dan may also belong here (cf. above, p. 353).

7, or as it is usually written y, occurs as final element in all
the forms of Ilokan dayto-y, in all the Ibanag forms with final
element ye (<*ya-£, cf. above, p. 331), and in Nabaloi ia-i,
Bagobo to-y ; it occurs followed by another element in Bagobo
to-i-yo, and in the double ligature ai.

A seems to occur as final element in Magindanao ini-a, and
in the double ligature ai.

Ai, a combination of the two ligatures a and i, occurs as final
element in all the forms of Ilokan dedi-ay, and in Nabaloi saidi-
ai) sadi-ai, and perhaps iai (cf. above, p. 353).

358

F. R. Blake,

[1906.

Interrogative Pronouns.

The interrogative pronouns are of three kinds :

a) The personal interrogative * who ? ' referring to persons ;

b) the neuter interrogative ' what ?' referring to things ;

c) the individualizing interrogative ' which ?' referring to
either persons or things.

In general the interrogatives consist of root pai'ticles, some of
which seem to be of adverbial origin, combined with prefixes
and suffixes similar to those of the demonstratives.

Case, in those interrogatives which are inflected, is indicated
by the prefixes. The plural is indicated sometimes by the
articular prefix, sometimes by reduplication, or by both together;
sometimes by the suffix.

Some of the interrogatives cannot be broken up into mono-
syllablic particles, but for the sake of completeness they are all
included in the discussion.

The forms of the interrogatives in the various languages are,
viz. :

PEBS. INTER.

NEUT. INTER.

now.

gen. obi.

Tag.

sg. sino

nino, kanino (sa) kanino ano

pi. si no si no

ninonino, (sa) kanikanino anoano

kanikanino

A

f ~\

Bis. (Ceb.)

kinsa, kinsalan

(kansa, kansalan sakansa)1 unsa, unsalan

Bis. (Hil.)

sin-o

sin-o, ni sin-o kay sin-o ano

kay sin-o, kalin-o kalin-o

kanin-o kanin-o

Bis. (Sam. -Ley.) sin-o

kanay, nin-o kanay, ano

sa kanay,

san kanay,

sin kanay

Bik.

sg. siisay

niisay kiisay ano

pi. saisay

naisay, kaisay kaisay

Pamp.

sg. nino

nino, ka-nino keno nanu

pi. dinino, deno

dinino, deno kareno

Pang.

sg. siopa, opa

opa ed siopa anto

pi. siopara

opara ed siopar^,

1 Little used.

Vol. xxvii.] Contributions to Philippine Grammar. 359

Ilok. asino, asinno, ania

asin

f dyano,
Igor. (Nab.) sepa, sepay •) ngaramto(i)

( nganto
Iban. sg. sinni nini takkuanini anni

/ takknarani
pi. dani, danirani dani 1 takkuarani-

( rani

Mag. tingin, antain ngain, antuna

Sulu sio, hisio uno

Bag. sadan andin

INDIV. INTEE.

nom. gen. obi.

Tag. sg. alin

pi. alinalin
Bis. hain

Bik. arin

Pamp. sg. insa,1 insanu1 ningsa, ningsanu kingsa, kingsanu

kaningsa, kaningsanu

pi. dingsa, dingsanu dingsa, dingsanu karingsa, karingsanu
Pang. dinan

Ilok. adinno, adino,

adin

Igor. (Nab.) chinan, tua, twai

Iban. sg. yasi nasi tasi

pi. danasi danasi (no oblique)

Sulu hadien

As the case forms of the interrogatives are comparatively
few, they will be analyzed in connection with the discussion of
the root particles.

A root particle no is found in all the forms of the personal
interrogative in Tagalog, Hiliguayna, Samaro-Leytean (except
kanay and derived forms), Pampangan, and Ilokan; in the
individualizing interrogative in Pampangan and Ilokan, and

1 The initial in in these forms is without doubt the article ing ; in this the most
common case form of the word, ng has been partially assimulated to the follow-
ing s, a change which has apparently not taken place in the other case forms.

360 F. R. Blake, [1906.

perhaps in the neuter interrogatives ano, uno. The glottal
catch in Bisayan -n-o is probably secondary. In Ibanag, no
occurs in the form ni, in the personal, and perhaps in the neuter
interrogative, the i of which is due to the influence of final i
of the individualizing interrogative yasi.

The nominative singular of the personal interrogatives derived
from no is made by prefixing the personal article si to the parti-
cle, except in Pampangan, where ni, ordinarily a genitive sign,
but also used for the nominative in Ilokan, is employed. The
initial a of Ilokan a-sino is perhaps due to the analogy of the
neuter interrogative ania ; asin is a shortened form of asino.
The doubling of the n in Ilokan asinno, Ibanag sinni, is pho-
netic.

The genitive singular is made with the case signs ni and ka.
Ni is prefixed directly to the particle no in Tagalo nino, Sam-
aro-Leytean nin-o, Pampangan nino, Ibanag nini ; in Hili-
guayna ni sin-o it is prefixed to the nominative. Ka is prefixed
to the genitive with ni in Tagalog, Pampangan, and Hiliguayna.
Hiliguayna kalin-o seems to be derived from kanin-o by dissimi-
lation, the first n being changed to the related sound I. Hili-
guayna also possesses a form made by prefixing kay, the oblique
of the personal article, to the nominative, viz., kay sin-o.

The oblique case in Tagalog is identical with the genitive
form with case sign ka, or is made by placing the oblique of
the definite article before this form; in Hiliguayna it is identi-
cal with the genitive forms beginning with k; in Pampangan
the oblique keno is probably a modification of ka-no with case
sign ka, based on the analogy of the oblique cases with initial
ke in the demonstrative and personal pronouns; in Ibanag the
oblique case sign takkua (cf. above, p. 347) is prefixed to the
genitive.

In Tagalog the cases of the plural are made by reduplicating
two syllables of the corresponding singular form. In the Pam-
pangan nominative-genitive plural there are two forms, viz.,
dinino, made by prefixing di, the root of the inclusive article,
to the singular nino, and deno, probably modified from di-no
after the analogy of the plurals of demonstratives with initial
de. The oblique is made by prefixing ka to deno. In the
Ibanag plural, dani consists of ni with prefixed plural pai'ticle

Vol. xxvii.] Contributions to Philippine Grammar. 361

\

da, danirani is a reduplication of this form, and the oblique
forms consist of the case sign takkua prefixed to these two
forms.

It is not impossible that the neuter interrogative ano is a com-
bination of no with an articular prefix a, but it may also be
explained as containing the root particle an. The u of Sulu
uno, if it is to be analyzed as ^^,-no, and is not simply a phonetic
modification of ano, is probably the same u which is used in
Batan as definite article.

No also occurs as the final syllable nu, no of the individual-
izing interrogative in Pampangan and Ilokan.

This stem no is perhaps identical with the particle no which
is the word for 'if,' in Ilokan, Ibanag, and Pangasinan, the
ideas of ' doubting ' and ' questioning ' being very similar.

A root particle sa occurs in all the forms of the personal
interrogative in Cebuan, Bikol, and Bagobo, 'the neuter inter-
rogative in Cebuan, and the individualizing interrogative in
Pampangan.

The initial k of the Cebuan personal interrogative forms
is probably the same which occurs as the sign of the nomi-
native in the demonstratives. The case in this pronoun seems
to be indicated not according to the usual rule by initial
inflection, but by the difference of the vowel after the initial,
i indicating the nominative, a the genitive-oblique. The par-
ticles in, an, un, which precede sa in Cebuan and Pampangan,
seem to consist of the articular particles i, a, u -j- the ligature;
this seems to be shown by the fact that the in of the Pampan-
gau forms is declined like the definite article. The final element
Ian, of the Cebuan forms is perhaps to be connected with Taga-
log alin, Bikol arin. The nu of insanu is to be connected with
the stem no. The case forms of the Pampangan individualizing
interrogative all consist of sa or sanu preceded by the proper
case form of the definite article, except in the oblique singular,
where additional forms are made by prefixing ka to the genitive,
following the analogy of the plural.

The forms of the Bikol personal interrogative are based on an
element isay, composed of sa with prefixed articular i and fol-
lowing ligature i. To this stem the case forms of the personal
and inclusive articles are prefixed in singular and plural respect-
ively, ka being equivalent to Cebuan ka (cf. p. 348).

362 F. R. Blake, [1906.

/

In Bagobo sa-dan, dan seems to be the genitive plural of the
third person, just as the din of an-din 'what?' is the genitive
singular (cf. below).

This inteiTOgative root particle sa is perhaps identical with the
adverbial particle sa which means 'perhaps' in Ilokan, and in
Tagalog is equivalent to the modal adverb sana which imparts
to verbs the idea of ' should, would.'

The forms of the personal interrogative in Pangasinan and
Nabaloi are clearly connected with Malay apa ' what T siapa
' who ?' Pa may be root particle and the prefixes articular ; o
of the Pangasinan forms being equal to Batan u\ se of the
Nabaloi forms, to sa-i, as in saidiai ' this ;' a of Malay apa, to
the a of Tagalog ana. The si of the Pangasinan and Malay
forms is of course the personal article, and it is also not impos-
sible that Nabaloi se may be some modification of si. The final
y of Nabaloi sepay is ligature. In Pangasinan the form with-
out si is used as genitive, and the oblique is made by placing
the oblique of the definite article ed before the nominative.
The plural is made by adding ra (<da) to the forms of the
singular.

The root particle of the neuter interrogatives in Pangasinan,
Ilokan, and Bagobo is an, and the same an is perhaps the root
particle of all the other neuter interrogatives except the Cebuan
and Sulu, and the forms ngaramto(i), nganto, ngain; Sulu uno,
however, may be a modification of ano (cf. above, p. 361).

The o, u of Tagalog, Bisayan, and Bikol an-o, Pampangan
nan-u, Nabaloi dyan-o, according to this view, is probably due
to the analogy of sino or some other interrogative form contain-
ing the particle no; the initial n of the. Pampangan form is
doubtless derived from nino; the initial dy (=*y] in Nabaloi
is probably articular. The final i of Ibanag anni, in which the
doubling is phonetic, is probably due to the same cause as the
final * in sinni.

In Pangasinan an-to and Bagobo an-din (cf. inter, sa-dan,
above) the second element is apparently the genitive of the pro-
noun of the third person, literally ' its what ?' just as in Nabaloi
the common word for ' what ' is ngaram-to ' its name ?' The
first part of Magindanao an-tu-na seems to be identical with
Pangasinan anto; na is perhaps identical with the genitive of
the definite article, the root particle in this case being followed

Vol. xxvii.] Contributions to Philippine Grammar. 363

by a double genitive. Ilokan ania is to be analysed in the
same way, as an -\- an element ia equivalent to the ya of the
third personal pronoun.

The stem an seems also to occur in some of the personal and
individualizing interrogatives, viz., in Magindanao antain, Pan-
gasinan di-n-an, Nabaloi chi-n-an.

This an is perhaps identical with the locative suffix an of
nouns and verbs.1

For Nabaloi ngaramto, cf. above; the i of ngaramto-i is
ligature ; nganto is contracted from nyaramto, m being assimi-
lated to the following t.

Tagalog alin (<*arin cf. above, p. 333), Bikol arin seems to
be based on a stem ar, the ending in being probably the same
as the in in Bisayan hain, and perhaps Magindanao tingin,
antain, ngain. This suffix in is different from the in of Taga-
log ak-in 'mine' (cf. below, p. 368), as this in would appear in
Bisayan and Bikol as on, un, in Magindanao as en (cf. above,
p. 331). The same stem occurs perhaps also in the Cebuan
forms kinsalan, kansalan, unsalan which may be contracted
from *kinsa-alan, etc., al (<*ar) being in this case combined
with a suffix an identical with that in Tagalog haan ' where ?'
which bears the same formal relation to hain as *alan does to
alin. This suffix an is doubtless ultimately identical with the
locative suffix an.

The Ibanag individualizing interrogative is clearly connected
with the interrogative adverb dasi, the element asi being the
same. This element asi forms its cases like the demonstrative
pronouns (cf. above, p. 356 f.).

The idea of ' which ?' is closely related to the idea of ' where ?'
as we see in the two practically synonymous expressions ' which
is the road ?' and ' where is the road ?' Hence it is not surpris-
ing that the two ideas are sometimes expressed by the same
word, as in Bisayan hain^ Nabaloi twa, twai, Sulu hadien, which
mean both ' which ?' and ' where ?'

As ' where ?' is practically an oblique case, we might expect
the initial syllables to be identical with the oblique case signs in
these individualizing pronouns, and so they seem to be. The

1 Cf . my paper, Analogies between Semitic and Tagalog, JHU. Circs. ,
No. 163, p. 65.

364 F. E. Blake, [1906.

ha of Bisayan ha-in, Sulu ha-dien is to be connected with the
oblique of Sulu definite article ha. This ha also occurs in Taga-
log in ha-an ' where ?' The in of Bisayan hain is difficult (cf.
above, p. 363). For the dien of Sulu hadien see below.

The di of Pangasinan di-nan, Nabaloi chi-nan (<*o?i-ncm),
Ilokan a-di-no, etc., and Sulu ha-di-en is probably identical
with the oblique articular form di. The nan of the Pangasinan
and Nabaloi forms is probably the interrogative stem an with
preceding ligature n, used to connect the prefix to the root
particle, as it is in the genitive of the demonstratives in Pan-
gasinan (cf. above, p. 356). The initial a of the Ilokan forms
is to be explained like the a of aslno; the doubling in adinno
is phonetic; adin is a shortened form of adino, like asin from
asino. The final no of these forms is probably identical with
the interrogative root particle no.

The en of Sulu hadien is probably connected with the demon-
strative particle en (cf. above, p. 353).

Nabaloi tua is perhaps identical with Cebuan tua ' be there ;'
.the i of twa-i is ligature.

The remaining interrogative forms, Samaro-Leytean kanay,
Magindanao tingin, antain, ngain, Sulu sio, hisio, are not clear.
Kanay contains, of course, .the case sign ka or k; its ending
suggests a comparison with the Bikol forms; in the oblique it
may be preceded by sa and san, respectively oblique and geni-
tive of the definite article, and by sin the sign of the indefinite
object. The Magindanao forms all have the ending in (cf.
above, p. 363) ; antain probably contains also the interrogative
stem an; ngain may be a combination of nga identical with the
ligature, and the suffix in. Sulu sio seems to contain the per-
sonal article si, which, however, is not used as such in Sulu;
while hisio has prefixed, in addition, the regular Sulu personal
article hi.

Personal Pronouns.

The personal pronouns are of three persons, first, second, and
third; each person has two numbers, a singular and a 'plural,
and the plural of the first person again distinguishes two series
of forms, an exclusive and an inclusive series (cf . above, p. 325) ;
the first person in some languages possesses also a dual which is

Vol. xxvii.] Contributions to- Philippine Grammar.

365

closely related morphologically to the inclusive plural ; there is
no distinction of gender even in the third person.

Case is indicated partly by prefixes similar to those of the
demonstratives and interrogatives, partly by using different root
particles and stems.

The personal pronouns consist of simple root particles, modi-
fied forms of root particles, and forms derived from the root
particles by the addition of prefixes and suffixes.

First Person.

The forms of the personal pronouns of the first person in the
various languages are, viz. :

nom.

gen.

obi.

Tag.

sg. ako

akin, ko, (nakin)1

sa akin

pi. ex. kami

amin, namin

sa amin

pi. inc. tayo

atin, natin

sa atin

du. kita, kata

ata, kanita; ta

sa ata, sa kanita

Bis.

(Ceb.) sg. ako

ako, nako, ko, ta

kanako, sa ako

pi. ex. kami

amo, namo

kanamo, sa amo

pi. inc. kita

ato, nato, ta

kanato, sa ato

du. kita

ato, nato, ta

kanato, sa ato

Bis.

(Hil.) sg. ako

akon, nakon, ko,

kanakon, sa akon,

ta

dakon, karakon

pi. ex. kami

amon, namon

kanamon, sa amon,

damon, karamon

pi. inc. kita

aton, naton, ta

kanaton, sa aton,

daton, karaton

Bis.

(Sam. -Ley.) sg. ako, si ako

akon, nakon, ko

sa akon, dakon

pi. ex. kami, si kami

amon, namon

sa amon, damon

pi. inc. kita, si kita

aton, naton, ta

sa aton, daton

Bis.

(Har.) sg. ako, akota

akon, nakon, ta

kanakun

Bik.

sg. ako

ko, niako, sako,

sako, sakoya

sakoya

pi. ex. kami

niamo, samo,

samo, samoya2

samoya, mi

pi. inc. kita

niato, sato,

sato, satoya

satoya, ta

1 Used only in the phrase a-nakin ' inquam.'

2 Written damoya, evidently a mistake, in San Augustfn.

366

F. E. Blake,

[1906.

nom.

gen.

obi

Pamp.

sg. ako, ko

ko, koo, ke, kee,

kanako, kako

da

pi. ex. ikami, ike,

mi

kekami, keke

kami, ke

pi. inc. ikamo, ita,

tamo,1 ta

kekatam (sic),

katamo, kata,

kekata

tamo, ta

du. ikata, kata, ta

ta, tee, too

kekata

Pang.

sg. siak, ak

ko, -k

ed siak

pi. ex. sikami, kami

mi

ed sikami

pi. inc. sikiti, sikatayo,

ti, tayo

ed sikiti, ed

iti, itayo

sikatayo

du. s;kata, ita

ta

ed sikata

Ilok.

sg. siak, ak

ko, -k

kaniak

pi. ex. dakami, kami

mi

kadakami

pi. inc. datayo, tayo

tayo

kadatayo

du. data, ta

ta

kadata

Igor. (Nab.

) sg. sikak, ak; nak,

ko, -k, na, ta

na

pi. ex. sikame, kame

me

V

pi. inc. sikatayo, tayo

tayo

Iban.

sg. sakan, sok, ak

ku, -k, ta

niakan, niok

pi. ex. sikami, kami

mi

n ikami

pi. inc. ittam

tarn

nittam

du. itta

ta

nitta

Bat.

sg. yakin, ako

niakin, ko

diakin

pi. ex. yamuen, kami

niamuen, namin,

diamuen

namuen

pi. inc. yaiatin,3 ta

Mag.

sg. saki, aku

ku, salaki, laki

salaki, sa salaki

pi. ex. salkami, kami

salkami, nami,

salkami, sa salkami

lekami

1 Written iamo, evidently a mistake, in Bergafio.

2 Perhaps a mistake for yatin, cf. yamuen. The forms given as genitive and oblique
of this pronoun evidently belong to the pronoun of the second person. The Batan
grammar from which these paradigms are quoted by Retana (cf. above, p. 323, ft. nt. 2)
is in manuscript, according to Retana (p. xl) probably a copy of the original. The
personal pronouns are given in the order, I9t sg., 1st exc. pi., 2nd sg., and then follows
a paradigm headed ' Plur. de yo inclusive,' but with genitive and oblique apparently
belonging to the second plural; no paradigm of the second plural is given. The copyist
has evidently mixed up the paradigms of the 1st inc. pi. and 2nd pi. The forms that
we should expect in I8t inc. pi. gen. and obi. are niatin, diatin.

Vol. xxvii.] Contributions to Philippine Grammar.

367

Mag.

Sulu.

Bag.

pi. inc. salkitanu, seki- salkitanu, tanu, salkitanu, sa salki-

sa salkitanu

tanu, tanu,

lekitanu
du. salkita, sekita, salkita, sekita,

ta

sg.1 aku, ku
pi. ex. kami

tanu

salkita, sa salkita

pi. inc. kita

sg. sakkan
pi. ex.* kami, ke, si kami de
pi. inc.* kita, si kita ta, nita

ta, lekita
ku, kaaku
kaamu, kannamu,

kanamu
kaatu
ko

kanakkan

kanami

kanita

The pronouns of the first person singular are practically all
based on one of the root particles ak, ko, or ta.

Ak is found uncombined as nominative in Pangasinan, Ilokan,
Nabaloi, and Ibanag: also in Tagalog, Bisayan, Bikol, Pam-
pangan, Batan, Magindanao, and Sulu nominative ako, aku: in
the ako, aku of the genitive and oblique cases in Bikol and Sulu,
and in Harayan ako-ta; in the Tagalog and Batan stem ak-in
and Bisayan stem ak-on, dk-o and their derivatives ; in the nomi-
natives, Pangasinan and Ilokan si-ak, Nabaloi sik-ak, Ibanag
s-ak-an, s-ok (cf. above, p. 336), Bagobo s-ak-kan, and the
oblique forms derived from them ; in the Magindanao secondary
stem aki in s-aki, etc. ; in Nabaloi n-ak.

The nominative ako may represent a blend of the two root
particles ak and ko, or it may be ak with an analogical o as in
ikao 'thou' (cf. below, p. 375), due to the influence of other
nominatives in o like tayo 'we,' kayo 'you.' In Harayan
ako-ta the genitive ta is added to this form, probably for the
sake of emphasis3 (cf. Nab. sikam 'thou,' p. 375,' below).

The Magindanao stem aki probably owes its final i to the
analogy of the forms of the exclusive plural kami, etc. The
n of Nabaloi n-ak is not clear; the form na, which seems to be

1 The nounpateJc ' slave ' is also used for ' I,' as in Malay.

8 These pronouns as given together in the paradigms as the equiva-
lent of nosotros ' we,' but there is probably the same distinction here as
in the other languages; for ke as exclusive plural cf. Pampangan.

3 Cf. my paper, The Bisayan Dialects, cited above, p. 127.

368 F. E. Blake, [1908.

employed principally before verbal forms beginning with k,
e. g. , na-kaama-munu 'I am killing,' is probably a modifica-
tion of nak, due to the fact that the final k of the pronoun and
the initial k of the verb coalesce.

The Tagalog, Batan, and Bisayan stems ak-in, ak-on are
made by adding the suffixes in and on to the root particle.
These suffixes, which are variants of the same original form (cf.
above, p. 331 f.), are found also in the passive, and in nominal
derivation. When applied to a noun they express the idea of
'like, similar,' e. g., Tag. uwakin 'cock black as a raven' from
weak ' raven,' polotin 'honey (pet name),' from polot 'honey;'
when applied to a verb they form passives that denote to be
dii-ectly affected by the action of the root, e. g., Tag. inumin
(inum) 'that which is drunk.' The meaning of the suffix in
the pronouns is perhaps nearer that of the verbal suffix, Tagalog
ak-in, for example, signifying ' that which is affected by me,
that which pertains or belongs to me.' In the Cebuan stem
dk-o, the final n was dropped, doubtless because it was regarded
as ligature. The ako, aku of the genitive and oblique forms in
Bikol and Sulu may be either the nominative ako, aku, or may
correspond to the Cebuan stem dk-o derived from ak-on. The
element ya of Bikol oblique s-ako-ya is not clear; it may be
the demonstrative root particle ya (cf. above, p. 352).

The stems of Ibanag s-akan, Bagobo s-akkan, are probably
identical, the doubling in the latter being phonetic, and consist
of the root particle with a suffix an, doubtless the same as the
nominal and verbal suffix an, which expresses the idea of place,
e. g., Tag. sagingan 'banana grove' from saging 'banana,'
inuman 'vessel (drinking place)' from inum 'drink.' These
forms would therefore mean literally 'the I place.'

The particle ko occurs uncombined as genitive in all the lan-
guages except Harayan, and apparently as nominative in Pampan-
gan and Sulu. In Pampangan it may be a shortened form of <iko.
In Sulu it is genitive used as nominative (cf. below, p. 388). The
-k of Pangasinan, Ilokan, Nabaloi, and Ibanag is a shortened
form of ko. The Pampangan form koo, ke and kee are modifi-
cations of ko (cf. below, p. 387).

The particle ta occurs uncombined as genitive in Bisayan
(except Samaro-Leytean), Nabaloi, and Ibanag. Pampangan da
is probably a modification of this particle. It is found also in
the Harayan nominative ako-ta (cf. above, p. 367).

Vol. xxvii.] Contributions to Philippine Grammar. 369

The pronouns of the exclusive plural of the first person are
practically all based on a particle mi, which occurs uncombined
as genitive in Bikol, Pampangan, Pangasinan, Ilokan, Nabaloi,
and Ibanag. It also occurs in Magindanao genitive na-mi, and
in the nominative lea-mi of all the languages, and its derivatives
with the various articular particles, viz., Pampangan i-kami,
Samaro-Leytean, Pangasinan and Nabaloi si-kami, Ibanag
si-kami, ni-kami, Ilokan da-kami, Magindanao le-kami, sal-
kami, sa sal-kami, and the oblique forms derived fr.om them in
Pampangan, Pangasinan, and Ilokan. The element ka of kami
is perhaps due to the analogy of other plural forms like kayo
'you,' kata 'we all, we two.'

In Tagalog, Bisayan, Bikol, Batan, and Sulu, there is appar-
ently a root particle am that forms in genera1! the same combi-
nation as the ak of the singular ; for example, in Tagalog, amin
corresponding to akin. It is very likely, however, that all these
forms with am are simply made on the basis of mi, after the
analogy of the forms of the singular. The final uen of the
Batan forms is simply the representation of in after the labial m,
cf. inumtien=Ta,galog inumin 'drink.'

Pampangan and Bagobo ke in ke, i-ke is not clear. In Bagobo
a particle de with the same vocalization, perhaps a modification
of the plural particle da, is used as genitive.

At the base of the forms of the inclusive plural, and of the
dual when it occurs, lies the particle ta, doubtless identical with
the ta of the singular, which is found uncombined as genitive
plural in Bisayan, Bikol, Pampangan and Bagobo; as genitive
dual in Tagalog, Cebuan, Pampangan, Pangasinan, Ilokan,
Ibanag, and Magindanao; and as nominative plural in Batan,
nominative dual in Pampangan, Ilokan, and Magindanao. The
Pampangan forms tee, too are secondary modifications of ta (cf .
below, p. 387). -For Pangasinan ti, cf. below, p. 371.

Ta occurs also, combined with articular prefixes only, in Taga-
log genitive dual ka-ni-ta and the oblique derived from it;
Pampangan nominative plural, Paugasinan nominative dual i-ta;
Ibanag dual forms i-tta, ni-tta, in which the doubling is pho-
netic ; and in Bagobo ni-ta, Ilokan da-ta, and the oblique forms
derived from them. It is found also in Tagalog genitive dual
ata and the oblique derived from it, where the initial a is prob-
VOL. xxvii. 25

370 F. R. Slake, [1906.

ably due to the analogy of the other genitive fonns of the first
person. For Pangasinan i-ti, cf. below, p. 371.

It occurs also combined with a following root particle of the
second person, which probably emphasizes the inclusive significa-
tion of the pronoun, the combinations meaning literally 'your we,'
' the ice of which you form a part,' in ta-yo, ta-mo, and ta-nu.
Ta-yo occurs as nominative inclusive plural in Tagalog, Ilokan,
and Nabaloi, and as genitive in Pangasinan, Ilokan, and Nabaloi.
It also occurs with various case indicating prefixes in the nomina-
tives Pangasinan i-tayo, Ilokan da-tayo, Pangasinan and Nabaloi
sika-tayo, and the oblique forms derived from the last two in
Ilokan and Pangasinan. Ta-mo occurs in Pampangan uncom-
bined as nominative and genitive, in the nominative ka-tamo,
in which ka is doubtless due to the analogy of plural and dual
forms with initial ka, like kayo ' you,' and the oblique ke-katam
in which the final o of katamo is dropped;1 it is found also
without o, in the Ibanag forms tarn, i-ttam, ni-ttam, the doubling
in the last two forms being phonetic. Ta-nu occurs only in
Magindanao : as all the forms of the inclusive phiral have a cor-
responding form in the dual without mi, it seems more likely
that nu was added to what are now the forms of the dual, but
which had originally both inclusive plural and dual meaning,
in order to differentiate between the two numbers, than that the
inclusive forms were made independently uponla, base tanu.

Ta also occurs in the plural and dual forms ka-ta and ki-ta :
ka is probably to be explained as the ka in ka-mi; ki-ta per-
haps represents a blend of ka-ta with i-ta, which occurs as nomi-
native dual in Pangasinan and Ibanag, nominative plural in
Pampangan. Kata occurs uncombined as plural nominative in
Pampangan, and as dual nominative in Tagalog and Pampan-
gan. It is found in combination with articular prefixes in Pam-
pangan nominative dual i-kata, oblique dual and plural ke-kata,
Pangasinan nominative dual si-kata and the oblique derived
from it. Kita occurs uncombined as nominative plural in Bisa-
yan, Bikol, Sulu, and Bagobo, and as nominative dual in Taga-
log and Cebuan : with articular prefixes in Samaro-Leytean and
Bagobo si kita; in Magindanao, where it is the basis of most

1 In Ilokan the pronouns of the first and second persons plural datayo,
tayo, dakami, kami, dakayo, kayo, often drop their final vowel in the
middle of a sentence: cf. Naves, op. cit., p. 49 f. ; H. W. Williams, op. cit.,
pp. 55, 56.

Vol. xxvii.] Contributions to Philippine Grammar. 371

of the forms of both dual and plural (cf . below, p. 373) ; and
in Pangasinan si-kiti (cf . below), and the oblique derived from it.

In Tagalog, Bisayan, Bikol, Batan, and Sulu plural there is
apparently a root particle at which is made on the basis of ta,
after the analogy of the singular, in the same way as the am
of the exclusive plural. The forms based on at are made in
the same way as those based on am. In Cebuan the forms
based on at are used also as dual, with, however, a different
accent.

The Pangasinan plural forms ti, i-ti, si-kiti seem to be modi-
fications of ta, i-ta, si-kita; the final i being probably due to the
influence of the exclusive forms mi, kami, etc.

The case indicating particles which are found in the pronouns
of the first person are, in Tagalog n, sa, kani, in Cebuan n, sa,
Jca(n), in Hiliguayna n, ka, ka(ii), d, sa, in Samaro-Leytean
si, n, d, sa, in Harayan n, ka(ri), in Bikol ni, s, in Pampan-
gan i, kan, k, k'e, in Pangasinan si, i, ed, in Ilokan si, da,
kani, ka, in Nabaloi si, sika, sik, in Ibanag si, s, i, ni, in Batan
i, yai(?), ni, n, di, in Magindanao se, s, le, I, sal,.na,sa, in
Sulu ka, kan, in Bagobo si, s, ni, kan, ka.

The prefixes i, ni, n, sa, ka, di, d, da, ke, ed, and the oblique
case sign s of Bikol have already been explained in connection
with the demonstratives (cf. above, p. 353 f.) ; kani is a combina-
tion of the two articular particles ka and ni. In the Bisayan
sign ka(n) it is difficult to say whether 'the n belongs to the case
sign or the following stem, i. e., whether forms like Cebuan
kanato are to be analysed as ka -f- the genitive forms like nato,
or as kan -J- the genitive forms like ato. Si is identical with the
si of the personal article, and s in Ibanag s-akan, Magindanao
saki, Bagobo s-akkan, is doubtless based upon it; Magindanao
se also is probably based on si, the e being due to the influence
of the case sign le. Pampangan k in kako is derived from ka,
a having coalesced with the a of ako. Nabaloi sika, sik in sika-
tayo, sik-ak are based upon a wrong division of forms like sikami,
a combination of the stem kami with the case sign si; in sikak
the a of sika coalesces with J,he a of ak. The apparent case sign
yai in Batan yaiatin, if it is not simply due to a typographical
error, is the result of the reduplication of the first syllable ya of
yatin, a combination of the stem atin and case sign i. Magin-
danao na of na-mi is identical with the genitive of the definite
article: since Magindanao has been strongly influenced by Ara-

372 F. R. Blake, [1906.

bic (cf. above, p. 322), the case sign le, I, which occurs in none
of the other languages, may be the Arabic preposition J li,
which is used to express a dative and sometimes a genitive idea ;
sal is a combination of this element with the case sign sa.

In the formation of the cases these particles are added, some-
times to a root particle, sometimes to a modified form of a root
particle or to a secondary stem, sometimes to another case
form. The forms of the nominative and genitive are often
without case sign.

In the nominative, Samaro-Leytean has si in all numbers
in the forms si ako, si kami, si kita : Pampangan has i in the
plural and dual forms i-kami, i-ke, i-ta, i-kamo, i-kata : Pau-
gasinan has si in all numbers in the forms si-ak, si-kami, si-kit i,
si-kata; sika in sika-tayo; and i in inclusive plural and dual in
the form i-ti, i-tayo, i-ta: Ilokan has si in the singular si-ak,
and da in the plural and dual forms da-kami, etc. : Nabaloi has
si in si-kame; sika, sik in sika-tayo, sik-ak : Ibanag has si in
si-kami; s in s-akkan, s-ok; and i in i-ttam, i-tta: Batan has
i(y) in all three numbers in y-akin, yaiatin (cf. above, p. 366,
ft. nt.), y-amuen: Magindanao has s in s-aki; se in se-kitanu,
se-kita; le in le-kitanu; and sal in plural and dual sal-kami,
sal-kitanu, sal-kita; the forms with le and sa/,are genitive-
oblique forms used as nominative: Bagobo has s in the singular
s-akkan; si in the plurals si kami, si kita. Nominative forms
without case sign, consisting of a root particle, a secondary stem,
or some modification of these, are found in all the languages,
Tagalog, Bisayan (except Samaro-Leytean), Bikol and Sulu hav-
ing only forms of this kind.

In the genitive in Tagalog, Bisayan, and Batan there are a
number of forms made on the basis of stems derived from the
root particles ak, am, and at by the addition of the suffix, in,
en, on. For the analogical character of the forms of the plural
cf. above, pp. 369, 371. These stems occur in Tagalog and Bisa-
yan both uncombined, and with case sign n, e. g., Hil. ak-on,
n-ak-on, in Batan only with the case signs ni or n, e. g.,
ni-amuen, n-amuen.

In Bikol the case signs of the genitive ni, s are prefixed in
the singular to ako, which is either nominative or identical with
Cebuan genitive ako., in the plural to the stems amo, ato, made
after the analogy of ako. In Sulu the stems aku, amu, atu, to

Vol. xxvii.] Contributions to Philippine Grammar. 373

which the ease signs ka, kan are prefixed, are susceptible of the
same two explanations as the Bikol forms. The double n in
kannamu is perhaps due to the analogy of forms like kan-nia
'his' (cf. below, p. 385).

In Magindanao in the singular the case signs I and sal are
prefixed to the secondary stem aki; in the plural and dual the
case signs are all prefixed to the stems kami, kitanu, kita except
in na-mi, where na is prefixed to the root particle. In Bagobo
the case sign ni is prefixed to the root particle in ni-ta. Geni-
tive forms without case sign, consisting of a root particle or
secondary stem, or of some modification of these, are found in
all the languages, Pampangan, Pangasinan, Ilokan, Nabaloi, and
Ibanag having only genitives of this kind.

In the oblique, Tagalog prefixes sa to the forms of the geni-
tive beginning with a, and in the dual also to the genitive with
case sign kani: Cebuan has two series of forms, one in which
sa is prefixed to genitive forms with initial a, one in which ka(n)
is prefixed to genitive forms (cf. above, p. 371) : Hiliguayna
has four series of forms, two like those in Cebuan, one in which
the case sign d is prefixed to forms with initial a, one in which
an additional case sign ka is prefixed to the forms with initial
d: Samaro-Leytean has two series of forms identical with the
Hiliguayna forms with initial sa and d: Harayan has apparently
only one series, identical with the series with initial k: the Bikol
oblique forms are identical with the genitives with initial s:
Pampangan makes its oblique singular by prefixing kan and k
to ako, probably the nominative, the oblique of the other mem-
bers by prefixing ka to the nominative with articular «, ka 4- i
becoming ke: Pangasinan prefixes ed to the nominative with
articular si: Ilokan makes its oblique singular by prefixing kani
to the root ak, the oblique of the other numbers by prefixing ka
to the nominative with initial da : Ibanag substitutes m, Batan
di, for the case signs of the nominative: in Magindanao, the
oblique case signs sal or sa sal are prefixed to the same stems as
in the genitive.

Second Person.
The forms of the personal pronouns of the second person are,

374

F. R. Blake,

[1906.

nom.

gen.

obi.

Tag.

sg. ikao, ka

iyo, mo (niyo)1

sa iyo

pi. kayo, (kamo)

inyo, ninyo

sa inyo

Bis. (Ceb.)

sg. ikao, ka

imo, nimo, mo

kanimo, sa imo

pi. kamo

inyo, ninyo

kaninyo, sa inyo

Bis. (Hil.)

sg. ikao, ka

imo, nimo, mo

kanimo, sa imo,

dimo, karimo

pi. kamo

inyo, ninyo

kaninyo, sa inyo,

dinyo, karinyo

Bis. (Sam. -Ley,

,) sg. ikao, ka, si ikao

imo, nimo, mo

sa imo, dimo

pi. kamo, si kamo

iyo, niyo

sa iyo, diyo

Bis. (Har.)

sg. ikao, kao, ka

imo, nimo, mo

kanimo

Bik.

sg. ika

mo, nimo, saimo,

saimo, simo

simo

pi. kamo

nindo

saindo

Pamp.

sg. ika, ka

mo, moo, me,

keka2

mee"

pi. ikayo, kayo,

yo, yoo, ye, yee

kekayo, keko

iko, ko

Pang.

sg. sika, ka

mo, -m

ed sika

pi. sikayo, siki,

y°

ed sikayo, ed siki

kayo, ki

Ilok.

sg. sika, ka

mo, -m

kenka

pi. dakayo, kayo

y°

kadakayo

Igor. (Nab.)

sg. sikam, ka

mo, -m

pi. sikayo, kayo

dyo

Iban.

sg. sikau, ka

mu, -m

nikau

pi. sikamu, kamu

nu

nikamu

Bat.

sg. imu, ka

nimo, mo

dimo

pi.3

ninio, nio

dinio

Mag.

sg. ka, saleka,

riengka, saleka,

salka, sa salka

seka

sa salka, leka, ka

pi. kanu, salkanu

salkanu, sa sal-

salkanu, sa salkanu

kanu, nu, niu,

lekanu

Sulu

sg. ekau, kau, nio

mn, kannio, kanio

pi. kamu

kaimu

Bag.

sg. kona, ka,

niko, no

kaniko, niko

si kona

pi. kio, ko, si kio

nio

kanio

1 Used only in the phrase a-niyo ' inquis.'

4 In place of the genitive and oblique forms, which are given elsewhere in the -text,
the forms of the pronoun of the first person singular are repeated in Bergano's paradigm.
3 Cf. above, p. 366, ft. nt. 2.

Vol. xxvii.] Contributions to Philippine Grammar. 375

The root particles upon which the pronouns of the second
person, singular and plural, are based, are ka, mo, yo, and nu.

Ka occurs uncombined as nominative singular in all the lan-
guages except Bikot and Sulu, and also as genitive in Magin-
danao. It is found also in the nominatives, Sulu and Harayan,
kau, kao. and with articular prefixes, Bikol and Pampangan i-ka.
Pangasinan and Ilokan si-ka, Tagalog and Bisayan i-kao, Sulu
e-kau, Ibanag si-kau, Nabaloi si-ka-m; in the oblique forms
Pampangan ke-ka, Pangasinan ed si-ka, Ilokan ken-ka, Ibanag ni-
kau; in all the forms with case signs in Magindanao. The forms
in o, «, probably owe this final element to the analogy of other
pronominal forms ending in o, u, like tayo, kayo. The final m of
the Nabaloi form is probably an added genitive element of the
same person, like the ta in Harayan ako-ta (cf. above, p. 367).
In the Bagobo genitive and oblique forms ni-ko, ka-ni-ko; ko
probably represents a modification of ka due to the analogy of
the genitive no. The particle ko which is found in Pampangan
and Bagobo in the plural forms ko, i-ko, ke-ko, is probably a
blend of the singular ka with the final o of the plural form
kayo, which o was felt as a sign of the plural. Similarly Pan-
gasinan ki in the plural forms ki, si-ki, ed si-ki, is probably a
blend of ka with the final i of plural forms of the first person,
viz., kami, sikiti, etc.

Ka occurs also in the plural stems ka-yo, ka-mo, ka-mi, which
are made up of two root particles of the second person. Kayo
is found uncombined as nominative plural in Tagalog, Pampan-
gan, Pangasinan, Ilokan, and Nabaloi, and with articular pre-
fixes in the nominatives, Pampangan i-kayo, Pangasinan and
Nabaloi si-kayo, Ilokan da-kayo, and the oblique forms derived
from them. Bagobo kio in kio, si-kio, is, perhaps, a modification
of kayo, due to the analogy of the genitive nio. Kamo occurs
uncombined in Bisayan, Bikol, Ibanag, Sulu, and in certain dia-
lects of Tagalog, and with articular prefixes in Ibanag si-kamu,
ni-kamu. Kami is found only in Magindanao both uncombined
and with articular prefixes. Bagobo kona in kona, si-kona, may
represent a metathesis of *ka-no. The ka of Pampangan ka-
tamo 'we,' and of the Pangasinan and Nabaloi case sign si-ka,
is an analogical element based on forms like ka-yo, ka-mo 'you,'
where kc(. is root particle.

376 F. R. Blake, [1906.

Mo occurs uncombined as genitive in'all the languages except
Magindanao and Bagobo. The genitive -m of Pangasinan,
Ilokan, Nabaloi and Ibanag is a shortened form of mo: it is
found also in Nabaloi nominative si-Jca-m (cf. above). Pam-
pangan moo, me, mee, are secondary modifications of mo (cf.
below, p. 387). Mo also occurs with case indicating prefixes in
the Bisayau genitives i-mo, ni-mo and the oblique forms derived
from them, in Bikol ni-mo, si-mo, sai-mo, Batan i-mu, ni-mo,
di-mo, and Sulu kai-mu; it is used, moreover, in the forma-
tion of the plural stems, second person Jca-mo (cf. above), first
person ta-mo (cf. above, p. 370).

Yo occurs uncombined as genitive plural in Pampangan,
Pangasinan, Ilokan, and Nabaloi (dyo < *yo, cf. above, p. 335).
Pampangan yoo, ye, yee are secondary modifications of yo (cf.
below, p. 387). With case indicating prefixes, yo is found in
the genitives, Tagalog singular, Samaro-Leytean plural i-yo,
ni-yo, Tagalog, Cebuan, and Hiliguayna plural in-yo, nin-yo,
and the oblique f OTTHS derived from them ; and ill the Batan
plural forms nin-io, din-io (io = yo] ; it is also used as
an element of the plural stems, second person ka-yo (cf. above),
first person ta-yo (cf. above, p. 370).

Batan, Magindanao and Bagobo genitive, Sulu, nominative
niu, nio, which is found also in Sulu ka-nio kan-nio, Batan
ka-nio, is probably to be analysed as case sign ni -(- yo, the form
representing either a different spelling of niyo, or a phonetic
variant, with y changed to the hiatus or glottal catch (cf . forms
with ia for iya, below, p. 381). Bikol nindo, saindo, are proba-
bly modifications of forms identical with Tagalog ninyo, sa inyo,
the d being due to the analogy of the corresponding forms of
the third person plural, viz., ninda, sa inda.

The particle no occurs uncombined as genitive plural in
Ibanag and Magindanao, and as genitive singular in Bagobo.
It is also used in Magindanao in the plural stem of the second
person ka-nu (cf. above) and as a final element of all forms of
the first person inclusive plural. It is not impossible that this
no is a modification of mo or yo, the n being due to the influ-
ence of other pronominal genitive forms with initial n.

The case indicating prefixes which are found in the pronouns
of the second person are, in Tagalog, i, n(i), in, sa, in Cebuan *,
n(i), in, sa,ka(ri), in Hiliguayna i, n(i], in, sa, ka(n), d(i), ka,

Vol. xxvii.] Contributions to Philippine Grammar. 377

in Sainaro-Leytean i, si, n(i) sa, d(i}, in Harayan i, n(i), ka(n),
in Bikol i, n(i), sa, s, «M, in Pampangan a, &a, in Pangasi-
nan si, ed, in Ilokan si, da, ken, ka, in Nabaloi s'i, in Ibanag si,
ni, in Batan i, n(i], d(i), in, in Magindanao s'e, sal, sale, le,
neng, sa, in Sulu e, ka, kan, i, in Bagobo si, ni, ka.

The prefixes * (of nom.), si, se, da, ni, sa, s, (of gen. -obi.),
sal, le, ka, ka(n), kan, ke, ed, have already been explained
(cf. above, pp. 353 f., 371). The prefix i in the Tagalog and
Bisayan genitive forms i-yo, i-mo is probably the same as the
prepositional i in Ilokan oblique of the definite article i-ti.
Whether the i in Batan nominative i-mu is this prepositional i
or the articular i of the nominative is difficult to say. The e
of Sulu ekau, though long, seems to be identical with articular i
(cf. pela 'how much ?' = Bis. pila). In the Tagalog, Bisayan,
Bikol, and Batan case signs n(i) and d(i) it is uncertain whether
the i is this prepositional i, or the final i of the case signs ni
and di, i. e. whether the forms niyo, nimo, diyo, dimo, in Tag-
alog, Bisayan, Bikol and Batan, are to be analysed as i-yo,
i-mo with prefixed case signs n and d, or as ni-yo, di-mo, etc.
The case signs n and d would of course be derived from ni and
di (cf. above, p. 354). Ilokan ken is probably contracted from
*ka-i-n with the two ligatures i and n. In Magindanao, sale is
a fuller form of sal; neng contains the ligature ng ; the e is an
indistinct vowel, as is shown by the spelling ngka,* and not the
result of the contraction of a diphthong; ne may, therefore,
represent a reduced form of nu, the genitive of the definite arti-
cle. The Tagalog, Cebuan, Hiliguayna, Bikol, and Batan prefix
in, which occurs both as initial and with additional prefixed case
signs, is not clear. It may be the prepositional i of iyo followed
by the ligature n, the prefix being thus similar to the prefixes
an, in, un, of the Cebuan and Pampangan interrogatives. In the
forms with compound prefix like nin-yo, din-yo, the i is of the
same ambiguous character as in forms like ni-yo, di-yo (cf.
above).

In general the case signs are. used in the formation of cases
as in the first person.

In the nominative Tagalog, Bisayan and Bikol have i in the
singular forms, i-kao, i-ka: Samaro-Leytean has si in the forms

1 Cf . Porter, Primer of the Moro Dialect, Washington, 1903, passim.

378 F. R. Blake, [1906.

si ikao, si kamo : Pampangan has i in both singular and plural in
the forms i-ka, i-kayo, i-ko: Pangasinan, Nabaloi and Ibanag
have si in both' singular and plural in the forms si-ka, si-kayo,
si-ki ; si-kam, si-kayo ; si-kau, si-kamu: Ilokan has si in singu-
lar si-ka; da in plural da-kayo: Batan has i, whether articular
or prepositional i is doubtful, in singular i-mu : Magindanao has
se and sale in the singular se-ka, sale-ka; sal in the plural sal-
kanu: Sulu has e=«, in e-kaw : Bagobo has si in the forms si
kona, si kio. Nominative forms without case sign, consisting
of a root particle or plural stem, or some modification of these,
are found in all the languages.

In the genitive in a number of languages secondary stems are
made by prefixing prepositional i and the particle in to the
root particles, viz., Tag. i-yo, in-yo, Bis. i-yo, i-mo, in-yo,
Bik. i-mo, in-do (cf. above, p. 376), Batan i-mo, in-io, Sulu
i-mu. In Tagalog and Bisayan these stems occur both uncom-
bined and with the prefixed case signs n, e. g., Tag. Inyo, n-inyo.
In the other, languages they are found only with case signs,
Bikol n(i), n, sa, s in nimo, n-indo, sa-imo, s-imo; Batan n(i),
n in nimo, n-inio, Sulu ka in ka-imu. In Magindanao the
case signs are prefixed to the nominatives ka and kanu; neng,
le, sale, sa sal to ka; le, sal, sa sal to kanu. In Sulu the case
signs ka and kan are prefixed in the singular to the form nio,
which though used as a nominative is formally a genitive ; in
the plural, ka is prefixed to the secondary stem imu. In Bag-
obo, ni is prefixed to the root particle in nio=ni-yo, and
ni-ko, in which ko is a modified form of ka (cf. above, p. 375).
Genitive forms without case, signs, consisting of a root par-
ticle either unchanged or in some modified form, are found in
all the languages, Pampangan, Pangasinan, Ilokan, Nabaloi,
and Ibanag having only forms of this kind.

In the oblique Tagalog prefixes the case sign sa to the geni-
tives with initial i; each of the Bisayan dialects has the same
series of forms based on the genitives as in the first person,
the analysis of the forms with initial kan, di, and kari being
ambiguous. In Bikol in the singular, as in the first person, the
oblique forms are identical with the genitives with initial s, in
the plural the case sign sa is prefixed to the secondary stem
indo. In Batan in the plural the case sign d is prefixed to the
secondary stem inio; the singular dirno has the case sign d(i).

Vol. xxvii.] Contributions to Philippine Grammar.

379

In Bagobo the case sign Jca is prefixed to the genitive in Jca-niko,
ka-nio; the genitive niko is also used as oblique. In Pampan-
gan, Pangasinan, Ilokan, Ibanag, and Magindanao the case
signs are prefixed to the nominative. In Pampangan, keka,
kekayo and keko are derived from the nominatives ika, ikayo
and iko by prefixing ka- kai becoming ke. Pangasinan forms
its oblique by prefixing ed to the nominatives with case sign si,
Ibanag changes this si to ni. Ilokan forms its singular by
prefixing ken to the nominative &a, its plural by prefixing ka to
the nominative with case sign da. In Magindanao, sal and sa
sal are prefixed to ka and kanu.

Third Person.

The forms of the personal pronouns of the third person are,
viz. :

nom. gen. obi.

Tag.

Bis. (Ceb.)

Bis. (Hil.)

sg. siya
pi.1 sila
sg. siya2
pi. sila
sg. siya2

sila

pi.

( sanda
Bis. (Sam. -Ley.) sg. siya

Bis. (Har.)

Bik.
Pamp.

Pang.

pi. sira
sg. tana

niya

nila

niya,2 iya,2 na

nila, ila

niya,2 iya

nila, ila

nanda, anda
niya, iya

nira, ira
ana, na

anda, nanda

niya
ninda

saiya

sainda

keya

pi. sanda

sg. siya

pi. sin da

sg. iya, ya, na, ne na, ne, nee, no,

noo

pi. ila, la, no da, de, dee, doo karela

sg. sikato to ed sikato

pi. sikara, ra, ira da, ra ed sikara

kaniya
kanila

kaniya,2 saiya,
kanila, saila
kaniya, sa iya, dia,

karia, sang iya
kanila, sa ila, dila,

sang ila

kananda, sa anda
kaniya, diya,

sa iya
dira, sa ira
kana, kanana,

kaniya
kananda

The forms of the plural may also be reduplicated, e. g., silnsila, kanikanila.
These forms are also written with ia instead of iya.

380

F. Ji. Slake,

[1906.

Ilok.

Igor. (Nab.)

Iban.
Bat.

Sulu
Bag.

sg. isu

na

kenkiiana, kaniana

pi. isuda, da, ida

da

kadakuada,

kaniada

sg. sikato, to

to

pi. sikai-a, si era,

cha, ra

era

sg. ya

na

sa

pi. ira

da

nira

sg. ya, sia

na, nia

dia,1 sia1

pi. sira, sa

nira, dara2

dira, sira,3 da,3 ra3

sg. salkanin,

salkanin,

salkanin,

sekanin, nin

co (

•< salkanin,
na (

kana ( „
•< salkanin
sa (

lekanin, nin

pi. silan, salkilan

kanilan, sa kani-

kanilan, sa kanilan

lan, nilan

sg. sia, nia

nia, kansia, kannia

pi. sila, nila

kansila, kanila

sg. kandin,

din

kandin

si kandin

pi. kandan,

dan

kandan

si kandan

The pronouns of the third person singular all appear to be
based on one of the particles ya, na, to, su, and nin. Ya and
na are probably identical respectively with the ligatures and
articular and demonstrative particles ya, na; to is probably
identical with the demonstrative particle to; su with the Batan
and Pangasinan article su, so; nin is perhaps the demonstrative
root particle ni -\- the ligature n.

Ya occurs uncombined in the nominative in Pampangan,
Ibanag, and Batan; with case indicating prefix in the nomina-
tive in Tagalog, Bisayan (except Harayan), and Bikol si-ya,
Batan and Sulu nominative sia (for si-ya}, Pampangan i-ya,

1 These case forms are headed ' Dat. ac. abl;' sia doubtless belongs to
ac.= accusative; cf. above, p. 341, ft. nt. 2.

2 Probably a mistake for da, ra, cf. oblique, and Pangasinan and
Nabaloi genitive. The use of a reduplicated form in the plural, how-
ever, is apriori quite natural, and is found in Tagalog (cf. above, p. 379,
ft. nt. 1), hence it is possible that the form is correct.

3 These forms are given under the head of ' Acus.'= accusative.

Vol. xxvii.] Contributions to Philippine Grammar. 381

Sulu nia (borrowed from the genitive), in the genitive in Taga-
log, Bisayan (except Harayan), and Bikol ni-ya, Batan and
Sulu nia (for ni-ya) , Bisayan (except Harayan) i-ya; and in
all the oblique forms in Tagalog, Bisayan (in Harayan only in
kani-ya], Bikol, Pampangan, and Batan. In the forms writ-
ten with ia for iya we have either a variant spelling, or the
intervocalic y has been changed to the hiatus or glottal catch
(cf. nio for niyo above, p. 376).

Na occurs uncombined as genitive in Cebuan, Harayan, Pam-
pangan, Ilokan, Ibanag, and Batan, and as nominative in Pam-
pangan. In Harayan it is the root particle of the pronoun in
all its forms with the exception of the oblique kani-ya. The
ta of the nominative tana is perhaps to be connected with the
demonstrative root particle ta. The initial a of the genitive
ana is probably due to the analogy of the plural form anda
(cf. below). Na also forms the basis of the Ilokan oblique
forms kenkua-na, kania-na. The Pampangan forms ne, nee,
no, noo, are probably secondary modifications of na (cf. below,
p. 387). It is not likely that no has any connection with the
particle nu, no of the second person.

To is found only in Pangasinan and Nabaloi, both uncom-
bined, and with prefixed case signs in Pangasinan and Nabaloi
sika-to, Pangasinan ed sika-to.

Su occurs only in Ilokan singular i-su with articular i, and
in isu, a prefix of the nominative plural (cf. below).

Nin forms the basis of all the forms of the pronoun in Magin-
danao. With this nin the particle din which forms the basis
of the pronoun in Bagobo is perhaps to be connected, the
initial d being due to the influence of the plural dan.

What Ibanag sa represents is doubtful. An original s seems
to be preserved in Ibanag only before i, otherwise becoming t
(cf . above, p. 333) ; hence sa can hardly be connected with the
articular sa of the other languages. It may be contracted
from *sia<*siya just as perhaps sakan 'I' from *si-akan,
though it is difficult to see why such a form should be used as
oblique.

The pronouns of the third person plural, with the exception
of Batan sa, which is perhaps identical with the inclusive arti-
cle of Bisayan and Bikol, are all based on one of the particles
da, ra,'la, which are perhaps ultimately identical.

382 F. R. Blake, [1906.

Da occurs uncombined as genitive in Pampangan, Pan-
gasinan, Ilokan, Nabaloi (cha < *da) and Ibanag; as nominative
in Ilokan ; and as oblique in Batan. Pampangan de, dee, doo
are secondary modifications of da (cf. below, p. 387). Bagobo
danin dan, kan-dan, si Jean-dan, is probably e?a-f-the ligature n.
.Da is found also with case indicating prefixes in Hiliguayna
and Harayan stem an-da, the Bikol.stem in-da, and in the Ilo-
kan forms isu-da, i-da, kania-da, kadakua-da, and in the com-
pound case sign ka-dakua of the last (cf. below, p. 384). It
also forms the first element of Batan genitive da-ra (cf. above,
p. 380, ft. nt. 2). The an and in of the stems an-da, in-da are
probably to be explained as articular a, i -\- ligature n, like the
prefixes an, in, un, of the interrogative pronouns (cf. above,
p. 361).

The particle ra occurs preceded by case signs ending in a
vowel in Samaro-Leytean, Pangasinan, Kabaloi, Ibanag and
Batan : la is found in Tagalog, Cebuan, Hiliguayna, Pampangan,
Magindanao, and Sulu under similar conditions; the final n of
the Magindanao forms is probably due to the analogy of the
singular. Ka occurs uncombined in Pangasinan genitive and
nominative, Nabaloi genitive, and Batan oblique, and in Batan
da-ra (cf. above). La occurs uncombined as nominative in
Pampangan.

It is difficult to determine with certainty the exact relation of
the three particles da, ra, la. Ha seems almost certainly to be
simply a phonetic modification of da, in all the languages in
which it occurs. This is indicated by the following facts. In
all these languages da and ra occur alongside of one another (in
Samaro-Leytean cf. the forms of the inclusive article), and as
the change of intervocalic d to r is a common phenomenon of
these languages (cf. above, p. 334), and as they seem to pre-
sent no change from r to d, the supposition lies near at hand
that the two are simply different forms of the same particle, da
being the more original form. This supposition is strengthened'
by the evidence of Ilokan, which does not change intervocalic d
to r, where the form corresponding to Pangasinan and Ibanag ira,
Nabaloi era. is Ida. According to this explanation the r of the
uncombined form ra of Pangasinan, Nabaloi, and Batan would
be due to the analogy of forms like ira, where the r is organic.

The particle la is perhaps a further phonetic modification of
this ra resulting from original da after a vowel. It is true that

Vol. xxvii.] Contributions to Philippine Grammar. 383

r resulting from intervocalic d is regularly retained in these
languages, but as words of very frequent use often suffer special
phonetic chanrges, it may be that in these pronouns the
secondary r became I just as original r became I in these lan-
guages (cf. above, p. 333). In this case the uncombined la of
Pampangan would be analogical like the uncombined ra.

It is also possible, however, to suppose that la is a particle
distinct from da. If so it can hardly represent an original la,
as in that case the intervocalic I would have been lost in Taga-
log and Sulu, but as in the languages where la is found I often
i-epresents an r of the other languages (cf. above, p. 333), the
original form of the particle was perhaps ra. If this supposi-
tion is correct, the ra in some of the forms may be referred to
the same original particle, as original r is usually retained in
these languages, but in just which forms, it would be difficult
to say.

The case indicating prefixes that are found in the pronouns
of the third person are in Tagalog si, n(i), ka(n), Cebuan si,
n(i], i, ka(n), sa, in Hiliguayna si, s, n(i), n, i, ka(n), d(i), ka,
sa, sang, in Samaro-Leytean si, n(i), i, ka(ri), d(i), sa, in Hara-
yan s, n, ka, ka(n], kan, in Bikol si, s, n(i~), n, sa, in Pampan-
gan i, ka, kare, in Pangasinan sika, i, ed, in Ilokan *',- isu, ken-
kua, kadakua, kanict, in Nabaloi sika, e, si, in Ibanag i, n(i),
(for the case sign of sa, cf. above, p. 381), in Batan si, n(i], d(i),
in Magindanao salka, seka, leka, sa, na, kana, si, salki, n(i),
ka(n], in Sulu si, n(i), ka, ka(n), kan, in Bagobo kan, si.

The signs i (of nom.), si, s (of nom.), sika, i (prepositional),
in, na, n(i), n, sa, d(i), ka, ka(n], kan, ed have already been
explained (cf. above, pp. 353 f., 371, 377).

The ambiguous signs are found in the following forms: n(i)
in niya, nia, nilan, nira, in Tagalog, Bisayan, Bikol, Ibanag,
Batan, Magindanao, and Sulu ; ka(n) in kaniya, kanilan in Taga-
log, Bisayan, and Magindanao, in kananda in Hiliguayna and
Harayan; d(i) in dia, dila, dirain Hiliguayna, Samaro-Leytean,
and Batan. I in Pampangan, Pangasinan, Ilokan, and Ibanag is
articular (cf . above, p. 353) ; in Bisayan, prepositional (cf . above,
p. 344) ; Nabaloi e is probably a modification of articular i due
to the following r; Hiliguayna sang is the genitive of the defi-
nite article ; Pampangan kare is probably borrowed from the
oblique plural of the demonstratives kareni, kareti, etc. ; Ilokan

384 . F. E. Make, [1906.

isu is the pronoun of the third person singular used as case sign ;
kenkua and kadakua contain the noun kua (cf. above, p. 347)
and the case signs ken (cf. above, p. 377), and kada, which is
derived by a wrong division from the oblique plural forms of
the first and second persons, e. g., ka-datayo, where the da
belongs to the stem and not to the oblique case sign; kania in
the forms kaniana and kaniada is a combination of the parti-
cles ka and ni with an a which is probably due to a wrong divi-
sion of the case signs containing kua, viz., kadaku-ana, kadaku-
ada. In Magindanao, kana is the genitive and oblique of the
definite article ; the ka of seka, leka and salka is probably due.
to a wrong division of forms like sal-kanu ' you ' where the ka
belongs to the pronominal stem and not to the case sign ; salki
may be borrowed in the same way from forms like sal-kita, ' we
two,' or it may be the case sign salka with the a changed to i
before la after the analogy of the other forms of the plural ; for
the elements se, le, sal, cf. above, pp. 371, 372.

In general the case signs are used in the formation of cases as
in the pronouns of the first and second persons.

In the nominative, Tagalog, Cebuan, Hiliguayna, Samaro-
Leytean, Batan, and Siilu, prefix si to the root particles ya and
la, ra in both singular and plural, in si-ya (sid), si-la, si-ra.
In Hiliguayna and Ha ray an plural the case sign s is prefixed to
the stem anda. In Bikol, si is prefixed to the root particle ya
in the singular, s, to the stem inda in the plural. In Pampangan,
i is prefixed to the root particle in singular i-ya, plural i-la.
In Pangasinan and Nabaloi in the singular, sika is prefixed to
the root particle in sika-to; in the plural both prefix sika and
i, e to the root particle da or ra, and Nabaloi makes an addi-
tional form by prefixing si to the form with case sign e. Ilokan
prefixes i to the root particle in singular i-su, plural i-da, and
also makes a plural form by prefixing isu to the root particle.
Ibanag has i in the plural i-ra. In Magindanao in the singular
the signs -seka, salka are prefixed to the root particle nin; in
the plural si and salki are prefixed to Ian, the root particle la
with an analogical n derived from the singular. In Sulu the
case signs si and ni are prefixed to the root particles in singu-
lar and plural, sia and nia being equivalent to si-ya, ni-ya: the
forms with ni are borrowed from the genitive (cf. below,
p. 388). In Bagobo the nominative of both numbers seems to

Vol. xxvii.] Contributions to Philippine Grammar. 385

be identical with the oblique; the nominative, however, may
take the additional case sign si.

Nominative forms without case sign are found in Harayan (tana
cf. above, p. 381), Pampangan, Pangasinan, Ilokan, Nabaloi,
Ibanag, Batan and Magindanao, consisting, with the exception
of Batan sa (cf. above, p. 381) of a root particle, two root par-
ticles (tana), or a modified form of a root particle.

In the genitive, Bisayan (except Harayan), as in the pronouns
of the second person, makes genitives by prefixing prepositional
i to the root particles, viz., i-ya, i-la. It is not impossible
that the genitive forms niya(nid), nila of Tagalog, Bisayan,
Bikol, Batan, Magindanao, and Sulu are based on this stem, the
case sign being n, but it seems more likely that they consist of
the case sign ni -(- the root particle. Hiliguayna, Harayan, and
Bikol prefix in the plural the case sign n to the stems anda,
inda. In Magindanao in the singular, the case signs leka, salka,
na salka, sa salka, are prefixed to the root particle nin; in the
plural the forms are based on Ian (cf. above), the case signs
being n(i), ka(>i), and sa ka(n}. Sulu kanila has the case sign
ka(n); in kan-nia, kan is prefixed to the genitive; in kan-sia,
kan-sila, to the nominative. Genitive forms without case
sign, consisting of a root particle or stem, or some modification
of a root particle, are found in all the languages except Tagalog,
Samaro-Leytean, Bikol and Sulu.

In the oblique, the Tagalog forms may be based either on the
genitive or the secondary stem with initial i, the case sign being
ka(ri). Each of the Bisayan dialects has in general the same series
of forms as in the pronouns of the first and second persons; the
forms with case sign sa are based on the secondaiy stem with
initial i; the forms kaniya, kanila, kananda have the case sign
ka(n} as in Tagalog; the forms with initial d and kar have the
case sign d(i) ; in Harayan, oblique forms are made from the
genitives na, ana by prefixing ka and kan respectively. In
Hiliguayna there is an additional series of forms made by
prefixing sang, the genitive of the definite article, to the
stem with initial i; in Samaro-Leytean, in addition to the regular
series, there is the form kaniya with case sign ka(n}. The
Bikol forms are made by prefixing sa to the stems with
prefixed i and in. Pampangan makes its singular by prefixing
ka to the nominative with articular i; in the plural we
VOL. xxvii. 26

386 F. R. Blake, [1906.

should expect *kela, but the form has become karela under
the influence of the demonstratives. In Pangasinan, ed is
prefixed to the nominative with case sign si. In Ilokan the
case signs are prefixed to the root particles, kenkua and kadakua
in the singular and plural respectively, kania, to both numbers.
In Ibanag the plural nira contains the sign n(i) ; for the singu-
lar sa cf. above, p. 381. Batan has the forms did, dira with
sign d(i], also sia and sir a identical with the nominative. In
Magindanao all the forms are identical with genitive forms,
except kana salkanin, in which kana salka is prefixed to the
root particle. In Bagobo kan is prefixed to the genitive. The
only oblique forms without case sign are Batan da and ra, which
are given as accusative forms (cf. above, p. 380, ft. nt. 3).
These forms may be really genitives, their presence under the
head of accusative being due to a mistake of the grammarian
or copyist (cf. above, p. 366, ft. nt. 2).

General Remarks on Case Formation.

Generally speaking, the inflection of pronouns in the Philip-
pine languages is initial, and is based on the forms of the arti-
cles. The nominative is often without case sign, especially in
the personal pronouns, but is also frequently indicated by initial
*', si, which are identical respectively with the i of the nomina-
tives, Ibanag «*, Pampangan i-ng, etc., of the definite article,
and si of the personal article ; it is also sometimes indicated by
s derived from si.

The genitive is most frequently characterized by an initial n.
As the ligatures na, n, are still often used between two nouns
that stand to one another in the relation of genitive and modi-
fied noun, e. g. Tag. bahay na bato, 'house of stone,' baro-n
kastila, 'a shirt of Spain, Spanish shirt,' it is very likely that
the genitive sign n is derived from them.

In the personal pronouns there are a number of genitive
forms without case sign. These forms, which are usually root
particles, are placed after the modified word, the genitive being
originally indicated simply by the postposition. Doubtless in
some primitive stage of the Philippine languages, any root par-
ticle might be thus used as a genitive, but in the course of
time, however, certain root particles were adapted to this use
(so with ko and mo of the first and second persons respec-
tively), others being excluded.

Vol. xxvii.] Contributions to Philippine Grammar. 387

In Pampangan the number of genitive forms of this kind has
been greatly increased by modifying the final vowel of the root
particle. There are five series of these forms, the endings
being a, o, e, oo, ee. Of the forms in a, all are unmodified
root particles except da (1. sg.) [cf. above, p. 368]. Of the
forms in o; ko, mo and yo are root particles; no (3. sg.) is modi-
fied from na on the analogy of these forms. The forms in e,
viz., ke, me, ye, ne, de, are all secondary; the e may have arisen
from the combination of a root particle ending in a, e. g. na
(3. sg.) with ligature i, and then have been extended to
the other forms. The forms in oo and ee are derived from
the o and e series respectively by repeating the final vowel.
The repetition seems to be emphatic, these forms being used
when the subject of the verb of which they are the agent, is
omitted.

This expression of the genitive by simple postposition has
been extended in Ilokan to the other pronouns, e. g. ti balay ti
tao, ' the house of the man,' iti atep toy a balay , ' the thatch of
this house,' so that the genitives of the various pronouns are
always identical with some form of the nominative.

The oblique case is never without a case indicating prefix, except
in the doubtful Batan forms da, ra (cf. above, p. 386). It is
usually characterized by an initial d, s, or k, derived respec-
tively from the oblique case forms of the articles di, sa, and Jca.
These oblique case forms are prepositional in character, but
have this peculiarity, viz., that they denote not only the case,
but also the idea of the definite, personal or inclusive article
before a following noun, For example, ' in the house ' is in
Tagalog not literally sa ang bahay, but simply sa bahay, the
oblique sa expressing both the idea of ' in,' and that of the
definite article.

It is quite common for forms which have oblique case signs
to be used as genitives, as, for example, in the Sulu personal
pronouns, and instances also occurs in which forms with geni-
tive case sign are used as oblique, for example, in the Cebuan
demonstratives and Ibanag personal pronouns.

The identity of nominative and genitive forms in Ilokan has
already been mentioned. Other instances of this identity are
found, especially in Pampangan and certain languages of the
Northern Group, and there especially in the inclusive article,

388 F. R. Blake, [1906.

and the plural of other pronouns. In the Sulu personal pro-
nouns the use of genitive forms in the nominative is probably
due to the fact that both nominative and genitive forms are
found after the case sign kan, e. g. kan-sia, kan-nia, thus giv-
ing rise to the idea that the forms like sia and genitive forms
like nia are equivalent.

Oblique case forms are found in the nominative in the Magin-
danao personal pronouns, the Cebuan demonstratives and
interrogatives with initial k, the Harayan demonstratives with
initial d, and the Bagobo pronouns of the third person. In the
Magindanao personal pronouns the case form's are very much
confused, some forms being used for all three cases. In the
demonstratives, the use of the oblique forms in the nominative is
probably due to the fact that certain adverbs of place, identical
with the oblique forms of the demonstratives, were used as
simple demonstratives, 'the there man,' or 'the man there,'
being used as the equivalent of 'this man,' (cf. the use of the
Cebuan adverb karon, above p. 351, ft. nt. 1.) The oblique case
signs thus introduced into the nominative became the regular
nominative case signs of the demonstratives, and in Cebuan,
were also extended to the interrogatives. The nominative
forms of the Bagobo pronouns of the third person are possibly to
be explained in the same way, inasmuch as pronouns of the third
person and demonstratives are very closely related.

The nominative forms given under the oblique in Batan are
probably used only in the accusative (cf . above, pp. 341, ft. nt. 2,
380, ft. nt. 3). The difficult Ibanag oblique sa (3. sg.) seems to
have a nominative case sign (cf. above, p. 381).

List of principal Elements used in the Formation of the Pronouns.1

a — 1) lig., cf. p. 338. 2) elem. of ligs. ay, ya, cf. p. 340. 3)
final elem. in dems. , cf . p. 357. 4) root part, of def . arts.
ang, an, cf. p. 341. 5) prefix in dems., cf. p. 354, and
dem. stems like ari, cf. p. 354 f. 6) elem. of anj and per-
haps of ad, ag. 7) perhaps dem. root part., cf. p. 352,
and root part of e.

ad — elem. of dem. stem adto, cf. p. 354 f.

ag — elem. of dem. stem agto, cf. p. 354 f.

• ' For elements of inc. art. not included here, cf. p. 348.

Vol. xxvii.] Contributions to Philippine Grammar. 389

akj — root part., 1. sg., cf. p. 367.

ak2 — sign of indef. object in Ceb., cf. p. 345.

al — cf. ar.

am — elem. of prons. 1. exc. pi., cf. p. 369.

an: — prefixed elem. in Ceb. inter, k-an-sa, cf. p. 361, and in the
stem anda, 3. pi., cf. p. 382.

an2 — prefixed elem. in nom. pi. of Iban. dems., cf. p. 356 f.

an3 — inter, root part., cf. p. 362.

an4 — probably suffix in Ceb. inters, ending in Ian, cf. p. 363.

an5 — suffix of prons. 1. pers., cf. p. 368.

ar — elem. of inters. Tag. al-in, Bik. ar-in, and perhaps of Ceb.
forms ending in Ian, cf. p. 363.

asi — stem of indiv. inter, in Iban., cf. p. 363.

at — elem. of prons. 1. inc. pi., cf. p. 371.

au — dem. root part., cf. p. 353.

ay, ai — 1) lig., cf. p. 338. 2) final elem. of dems., cf. p. 357.

bbad — final elem. of Bag. numeral 'one,' cf. p. 345.

chi — Nab. form of di.

d, — obi. case sign in Tag. dems., cf. p. 354; in prons. 1. pers.,
cf. p. 371; and in Bat. du, cf. p. 344: [cf. also d(i)~\.

d2 — nom. case sign in dems., cf. p. 354.

da, ra, la — 1) root part., 3. pi., cf. p. 381. 2) elem. of pi. def.
art. in Pang., cf. p. 342. 3) root part, of inc. art. in Ilok.
and Iban., elem. of inc. art. in Pang., cf. p. 348. 4) plu-
ral elem. of dems. in Pamp., Pang., and Iban., cf. p. 354.
5) elem. of prefixes dag and day. 6) prefix of gen. pi. of
Pang, dems., cf. p. 356. 7) prefix in Ilok. prons., 1. pi.,
du., 2. pi., cf. pp. 372, 378. 8) elem. of pers. inter, in
Pang., Iban., and Bag., cf. pp. 360, 361, 362, and of indiv.
inter, in Iban., cf. p. 363. 9) perhaps elem. of dday. —
Pampangan da gen. of pron. 1. sg. is not identical with
this da, but a modification of £«4.

dday — final elem. of Iban. numeral 'one,' cf. p. 345.

dag — prefix of pi. in Ilok. def. art. and dems., cf. pp. 342, 354.

day — prefix of dems. sg. in Ilok., cf. p. 356.

de, — gen. of pron. 1. exc. pi. in Bag., cf. p. 369. — Pamp. de
gen. of pron. 3. pi. is not identical with this de, though both
are probably modifications of da.

de2 — prefix of Pamp. inter, pi. de-no, cf. p. 360.

dij — 1) obi. of def. art. [Nab. chi], cf. p. 343, and of pers. art.,
cf. p. 347. 2) obi. case sign in Tag. dems., cf. p. 353.

390 F. E. Blake, [1906.

3) elem. of indiv. inter, in Pang., Ilok., Nab., and Sulu,

cf. p. 364. 4) obi. case sign in Batan prons. 1. pers., cf.

p. 371.— [cf. also <?(*)].
di2 — dem. root part., cf. p. 352.
di3 — 1) root part, of pi. of def. art. in Pamp., cf. p. 342.

2) root part, of inc. art. in Pamp., and elem. of inc. art.

in Pang., cf. p. 348. 3) prefix of pers. inter, pi. in Pamp.,

cf. p. 360. — perhaps a modification of da.

di4 — nom. case sign in Ceb, and Har. dems., cf. p. 354 ( = di,).
d(i) — obi. case sign, maybe dt or di^ in prons. 2. and 3. pers.,

cf. pp. 377, 383.
din — in Bag., root part., 3. sg., and final elem. of neut. inter.,

cf. pp. 381, 362.
ding — in Pamp., stem of pi. of def. art., and elem. of pi. of

indiv. inter., cf. pp. 342, 361.
e — nom. of def. art. in Nab., cf. p. 341. — e in Sulu ekau is

identical with i^ cf. p. 377.
ed — in Pang, only: ]) obi. of def. and pers. arts., cf. pp. 343,

346 f. 2) obi. case sign of all other pronouns, cf. pp. 347,

356, 362, 373, 379, 386.
ent — 1) dem. root part., cf. p. 353. 2) probably elem. of Sulu

inter, hadi-en, cf . p. 364.
en2 — cf. uen.

et — obi. and gen. of def. art. in Har., cf. pp. 342, 343.
g — pluralizing infix in Ilok. dems, cf. pp. 354, 356.
go — final elem. of Bag. dems., cf. p. 352.
ha — 1) obi. of def. art. in Sulu, cf. p. 343. 2) initial elem. of

indiv. inters, in Bis. and Sulu, cf. p. 364.
hi — in Sulu only, pers. art. and prefix of pers. inter., cf. pp. 346,

364.
ij — 1) Kg., cf. p. 338. 2) elem. of ligs. ay, ya, cf. p. 340. 3) final

elem. in Pang. def. art., cf. p. 342, and perhaps in the

forms e, ne of the def. and pers. art. in Nab. (cf. pp. 341,

343) ; in the forms of the numeral ' one ' in Iban. ; cf . p. 345 ;

in kay, obi. of pers. art. in Tag. and Bis., in the inc.

art. in Hil., cf. p. 348; in the dems., cf. p. 357; in the

following inters., viz., Bik. and Samaro-Leytean pers.

inter., cf. pp. 361, 364, Nab. sepa-i, twa-i, ngaramto-i;

followed by n in the case sign ken, and perhaps in Pang.

nen (cf. p. 346). 4) root part, in def. arts, i, in, ing, cf.

Vol. xxvii.J Contributions to Philippine Grammar. 391

p. 341 ; and pers. art. i and perhaps hi, cf . p. 346. 5) artic-
ular prefix or nom. case sign in Ilok. def. art., cf. p. 342;
in the forms of the numeral ' one.' cf. p. 345; in the dems.,
cf . p. 353 ; in the Bik. inter, stem isay, cf . p. 361, in
Iban. indiv. inter., cf. p. 363, and in Nab. inter. sepa(y] if
it is equivalent to *sa-i-pa(y], cf. p. 362; in the pers. prons.
of all persons, cf. pp. 371, 372, 376, 377 f., 383, 384.— The
root part, ya, ia, 3. sg. is perhaps a combination of a with
this prefix. 6) elem. of inv

ia — prepositional prefix or obi. and gen. case sign in Ilok. def.
art. and dems., cf. pp. 344, 353 f. , 356; and in prons. 2. and
3. pers., cf. pp. 377, 378, 383, 385: elem. of in^

ia — cf. y«.

id — elem. of dem. stem idto, cf . p. 354 f .

in, — prefix in inters., Ceb. k-in-sa, Pamp. insa (inking), cf.
p. 361, and in the stem inda, 3. pi., cf. p. 382.

in2 — prefix in stems inyo, indo, 2. pi., cf. pp. 376, 377.

in3 — suffix of gen. and obi. forms, prons. 1. pers. (=on, un,
and en^, cf. p. 372.

in4 — suffix of inters., cf. p. 363. — not identical with in3.

k, — nom. case sign in Ceb. dems. and pers. inter., cf. pp. 354, 361.

k2 — obi. case sign in inc. art., cf. p. 348; in Pamp. king, cf.
p. 343 f . ; in Bik. ki, cf . p. 346 ; in Har. k-agto, cf . p. 355 ;
and in Pamp. k-ako, cf. p. 373.

ka: — root part., 2. sg., cf. p. 375. •

ka2 — 1) obi. and gen. of def. art. in Bag. 2) obi. and gen. case
sign in def. art., cf. pp. 342-344; in pers. art., cf. p. 346;
in the inc. art., cf. p. 347 f. ; in dems., cf. p. 353; in the
inters, in Tag., Hil., and Pamp., cf. p. 360, and probably
in Sam. -Ley. kanay, cf. p. 364; in the prons. of all three
persons, cf. pp. 371, 376 f,, 383. 3) elern. of ka(ri), kan,
kay, ken, kana, kani, kare and takka. — The ka of the inc.
art. in Ceb. does not belong here, cf . &2 and kat.

ka3 — elem. of the Mag. case signs salka, seka, leka, cf. p. 384,
and of sika.

ka4 — in Ceb., obi. of inc. art., and in Bik., obi. case sign of pi.
of pers. inter., cf. pp. 348, 361.

kan — 1) obi. and gen. of the def. art., cf. p. 343 (bis); of pers.
art., cf. p. 346. 2) obi. and gen. case sign in pers. prons. in

392 F. R. Blake, [1906.

Har., Pamp., Sulu, and Bagobo, cf. pp. 371, 377, 383.
3) in Nab. obi. sikan, cf. p. 346.

ka(n) — obi. and gen. case sign, may be ka or kan, in the pers.
prons. of all three persons, cf. pp. 371, 376 f., 383.

kana — in Mag., gen. and obi. of def. art., and gen. case sign in
prons. 3. sg., cf. pp. 343, 344, 383.

kani — 1) gen. and obi. of pers. art. in Mag., cf. p. 346. 2) obi.
case sign in Ilok. kaniak, cf. p. 373. 3) elem. of case
sign kania in Ilok. prons. 3. pers., cf. p. 384.

kare — obi. case sign in Pamp. kareno, karela, cf. pp. 360, 386.

kay — 1) obi. of pers. art. cf. p. 346. 2) obi. and gen. case sign
in Hil. pers. inter., cf. p. 360.

ket — obi. case sign in Pamp. inter, keno, cf. p. 360.

ke, — nom. pron. 1. exc. pi. in Pamp., and Bag., cf. p. 369.
—Pamp. ke, 1. sg. does not belong here, being a modifica-
tion of ko, cf. p. 387.

ken — in Ilok., obi. of pers. art., cf. p. 346, and obi. case sign of
prons. 2. and 3. sg., cf. pp. 379, 3" 84.

key, xey — elem. of numeral ' one ' in Pang, and Nab. respect-
ively, cf. p. 345.

ki — in Bik., obi. of pers. art., cf. p. 346, and obi. case sign in
pers. inter., cf. p. 361. — probably a modification of ka^; not
identical with Pang, ki, 2. pi., which seems to be a modi-
fication of kat, cf. p. 375.

king — in Pamp., obi. of def. art. and obi. case sign of indiv.
inter., cf. pp. 343 f., 361.

ko — root part., 1. sg., cf. p. 368. — ko, 2. pi. does not belong
here, being probably a modification of &«,, cf. p. 375.

kua — a noun meaning 'possession' used in Iban. in the obi. case
sign takkua, and in Ilok. in the obi. of the prons. 3. sg.
and pi., cf. p. 384.

la — root part., 3. pi., cf. da.

le, 1 — gen. case signs in Mag. pers. prons., cf. pp. 372, 373, 378;
elem. of Mag. case sign leka, cf. p. 384, and of sale, sal.

ma — dem. root part., cf. p. 353.

mey — prefix of Ilok. numeral ' one,' cf. p. 345.

mi — root part., 1. exc. pi., cf. p. 369.

mo-^root part., 2. sg., cf. p. 376.

n1— 1) lig., cf. p. 338. 2) final elem. in def. art., cf. p. 341 ff. ;
in pers. art., cf. p. 346; in the dems., cf. p. 357; perhaps

Vol. xxvii.] Contributions to Philippine Grammar. 393

in prons. 3. pi. in Mag. and Bag., cf. p. 382 (bis). 3)
between prefix and root part, in the gen. of Pang, dems.,
cf. p. 356; in indiv. inters, in Pang, and Nab., cf.
p. 364; in the elems. an^ inlt unt. 4) in case signs kan,
ken.

n2 — gen. case sign in inc. art., cf. p. 348; in dems., cf. pp. 353,
354; in Iban. indiv. inter., cf. p. 363; in prous. 1. and
3. pers., cf. pp. 371, 372, 383, 385; and in all probability
in Bat. nu.

na,— 1) lig., cf. p. 338. 2) gen. of def. art., cf; p. 342. 3)
root elem. of gen. forms nan, nang, ka-nan, and perhaps
of ne (na-\- f) and nen (na -\-i-\-n) of def. and pers. arts.,
cf. pp. 343, 346; and of nay in Hil. inc. art., cf. p. 348,
4) final elem. of Tag. inc. art., cf. p. 347. 5) dem. root
part., cf. p. 353, and root part., 3. sg., cf. p. 381. 6) gen.
case sign of Pang, dems., cf. p. 356, and of Mag. prons.,
1. and 3. pers., cf. pp. 373, 385. 7) final elem. of inter.
antuna, cf. p. 362. 8) in kana.

na2 — in Bik. and Ceb, gen. and obi. of inc. art. and gen. case
sign of pi. of pers. inter., cf. pp. 348, 361; and final elem.
of inc. art., cf. p. 348.

ne — 1) gen. of def. and pers. arts in Nab., cf. pp. 343, 346. 2)
gen. of pron. 3 sg. in Pamp., cf. p. 381. 3) elem. of Pang,
gen. of pers. art. nen, cf. p. 346.

neng — gen. case sign of pron. 2. sg. in Mag., cf. p. 378.

nij — 1) gen. of pers. art., cf. p. 346. 2) gen. case sign in inc.
art., cf. p. 347 f . ; in dems., cf. pp. 353, 354; in the pers.
inter, in Tag., Sam. -Ley., Bik., Pamp., and Iban., cf. pp.
360, 361 ; in prons. 1. and 2. pers., cf. pp. 371, 372, 373, 377,
378. 3) obi. case sign in Ceb. dems., cf. p. 353, and in
Iban. prons. 1. and 2. pers. 4) root element of gen. forms
nin, ning of def. art., cf. p. 343. 5) final elem. in Hil.
inc. art., cf. p. 348. 6) in kani.

ni2 — dem. root part., cf. p. 352, and perhaps root part, of Mag.
nin, cf. p. 380.

ni3 — elem. of Iban. inters. = no1, cf. pp. 360, 362.

n(i) — gen. case sign, may be n1 or m,, in prons. 2. and 3. pers.,
cf. pp. 377, 378, 383, 385.

ning — in Pamp., gen. of def. art. and gen. case sign of indiv.
inter., cf. pp. 340, 361.

394 F. R. Blake, [1906.

no, — inter, root part., cf. p. 359 f.

noa — root part., 2. pers., cf. p. 376. — Pamp. gen. no, 3. sg.,

does not belong here, being a modification of na^ cf . p. 387.
11 u — in nu, ka-nu of def. art. in Bat. and Mag., cf. p. 342 ff.
ng — 1) lig., cf. p. 338. 2) final elem. in forms of def. art., cf.

p. 341 ff. ; in case signs identical with def. art. in Pamp.

indiv. inter., cf. p. 361. 3) in Mag. ne-ng-ka, cf. p. 377.
nga — lig., cf. p. 338.
on, un — suffix of gen. and obi. forms, prons. 1 pers. (=*ws and

en,), cf. pp. 372, 373.
pa, — obi. of def. art. in Sulu, cf. p. 343.
pa2 — perhaps inter, root part, in Pang, and Nab. pers. inter.,

cf. p. 362.

ra — root part., 3. pi., cf. da.
ri — = di intervocalic.

ro — final elem. of numeral 'one' in Bik., cf. p. 345.
s, — 1) obi. case sign in the articular forms sin, sun, cf. pp. 343,

344, 346; in the signs of indef. obj. sin, sing, cf. p. 345; in

the dems., cf. pp. 353, 354. 2) gen. and obi. case sign in

prons. 1. and 2. pers. in Bik., cf. pp. 371, 372, 377, 378.
sa — nom. case sign in Ceb. sini, cf. p. 354 ( = £,).
83 — nom. case sign in inc. art. sa in Ceb, Bik., and sanday in

Hil., cf. p. 348; in pers. prons. 3. pi., Bat. sa (?), Hil. and

Har. sanda, Bik. sinda, cf. pp. 381, 383, 384; and in prons.

1. sg., cf. p. 371.
sa, — 1) obi. and gen. cf def. art., cf. pp. 342, 343. 2) root

elem. of forms san, sang of def. art., cf. p. 343. 3) obi.

case sign in dems., cf. p. 354; in pers. inter, in Tag.,

Ceb., and Sam. -Ley., cf. p. 358; in pronouns of all three

persons, cf. pp. 371, 373, 376 f., 378 f., 383, 385 f. 4) elem.

of Mag. case sign sale, sal.
saa — 1) root elem. in Pang. def. art., nom. sg. say. 2) nom.

case sign in pi. of def. art., and in incl. art. in Pang., cf.

pp. 342, 348; in dems. in Pang, and Nab., cf. pp. 356, 357;

and perhaps in the Nab. pers. inter., cf. p. 362.
sa3 — root part, of numeral 'one,' cf. p. 345.
sa4 — inter, root part., cf. p. 361.
sa6 — 1) nom. of inc. art. in Ceb. and Bik., and of pron. 3. pi. in

Bat., cf. pp. 348, 381. 2) in Bik., nom. case sign of inc.

art. and of pi. pers. inter., cf. pp. 348, 361. 3) final elem.

of inc. art., cf. p. 348.

Vol. xxvii.] Contributions to Philippine Grammar. 395

sa6 — nom. case sign in Ceb. sa-ini, cf. p. 354 (=sa1).

sa7 — obi. of pron. 3. sg. in Iban., cf. p. 381.

sale, sal — case sign in Mag. pers. prons., cf. pp. 371 f., 377, 383.

san — in Sam. -Ley. only; 1) gen. of def. art. 2) obi. case sign in

pers. inter. Cf. pp. 343, 364.
sang — in Hil. only; 1) gen. of def. art. 2) obi. case signs in

prons. 3. sg. and pi. Cf. pp. 343, 385.

S6j — case sign in Mag. pers. prons., cf. references under sale.
se2 — initial elem. of pers. inter, in Nab., cf. p. 362.
si — 1) nom. of pers. art., cf. p. 346. 2) nom. of def. art. and

elem. of gen. nin-si in Bik., cf. pp. 341, 343. 3) nom.

case sign in inc. art., cf. p. 348; in pers. inter, in Tag.,

Hil., Sam. -Ley., Bik., Pang., Ilok., Iban., and Sulu, cf.

pp. 360, 361, 362, 364; in prons. of all three persons, cf.

pp. 371, 372, 377, 378, 383, 384 f. 4) initial elem. of sikan,

obi. of pers. art. in Nab., cf. p. 346. 5) final elem. in Hil.

inc. art., cf. p. 348. 6) elem. of case sign sika, sik.
sin — 1) gen. of def. art. in Sulu, cf. p. 343. 2) sign of indef.

object, and obi. case sign of pers. inter, in Sam. -Ley., cf.

pp. 345,' 364.
sika, sik — nom. case sign in Pang, and Nab. prons. 1. and 3.

perg., cf. pp. 371, 372, 383, 384.
so, su — 1) nom. of def. art., cf. p. 341. 2) in form tu in Iban.,

as sign of indef. obj., cf. p. 345. 3) root part., 3. sg. in

Ilok., cf. p. 381. 4) initial elem. of Nab. dems. suta, sutan,

cf. p. 357. 5) nom. case sign in Pang, dems., cf. p. 356.
t — in Iban. only; equivalent to s1? cf. p. 333: obi. case sign in

Iban. dems. and iridiv. inter., cf. pp. 354, 363.
ta, — dem. root part., cf. p. 353, and elem. of Har. tana, cf.

p. 381.

ta2 — root part., 1. sg. pi., and du., cf . -pp. 368, 369 ff.
ta3 — in Iban. only; equivalent to sa^ cf. p. 333. 1) obi. of def.

and pers. arts., cf. pp. 343, 347. 2) elem. of obi. case signs

takka, takkua.
ta4 — gen. and obi. of def. art. in Bag., cf. p. 343; probably not

identical with tas.
takka — in Iban. only; obi. case sign in pi. of dems., cf. pp. 354,

356 f.
takkua — in Iban only; 1) obi. of pers. art., cf. p. 347. 2) obi.

case sign in pers. and inc. arts., cf. pp. 347, 348; in pi. of

dems., cf. p. 356 f. ; and in pers. inter., cf. p. 361.

396 Blake, Contributions to Philippine Grammar. [1906.

ti — root part of def. art. in Ilok., cf. p. 341, and of Pamp. dem.
iti, cf. p. 353. — Pang, ti, gen. 1. inc. pi., does not belong
here, being probably a modification of ta^ cf. p. 387.

to — 1) dem. root part., cf. p. 352. 2) root part., 3. sg., in Pang,
and Nab., cf. p. 381. 3) prefix of gen. sg. in Pang. dems. ,
cf. p. 356. 4) elem. of inters, anto, antuna, cf. p. 362.

tu — sign of indef. obj. in Iban. = so, su, cf. p. 345.

u — 1) root part, of def. art. in Bat. and perhaps in Nab. sun,
cf. pp. 341, 342, 344. 2) articular elem. in Bis. numeral
usa, cf. p. 345; in the dem. stems uri, ura, cf. p. 354; and
perhaps in the Pang, inter, stem opa, cf. p. 362, and the
Sulu inter, uno, cf. p. 361. 3) elem. of un^

uen — suffix of gen. and obi. forms, prons. 1. pers. (=ins and
on, uti), cf. p. 369.

unj — articular elem. in Ceb. inter, unsa, cf. p. 361.

un2 — cf. on, un.

y— cf. v

ya, — 1) lig. in Pang., cf. p. 338. 2) nom. of def. art. in Har.,
cf. p. 341. 3) dem. root part., cf. p. 352 f. 4) root part.,
3. sg., cf. p. 380 f. 4) root elem. of Bag. def. art. yan, cf.
p. 341, and in Iban. dem. suffix ye, cf . p. 352, 357. 5) final
elem. of Ilok. ania, cf. p. 363.

ya, — suffix of prons. 1. pers. in Bik., cf. pp. 368, 369, 371.

yo — root part., 2. pers., cf. p. 376.

II.

IV.

V.

VI.

VII.

VIII.

Notes on some Palmyrene Tesserae. — By HANS H. SPOER,
Ph.D., Jerusalem. Syria.

For the opportunity to describe Nos. 1 and 2 I am indebted to
Mr. Herbert Clark of Jerusalem. The remainder are in my own
possession.

I. This tessera is a quadrangular oblong.

Obverse. Near the upper margin the inscription

A well-known name. Below it is a sun in splendor, sur-
rounded by circles and upturned crescents.
Reverse. The inscription runs the long way:

»rrv

Yarhai. Kalba.

is a name of frequent occurrence. frO^D occurs here,
so far as I can discover, for the first time in a Palmyrene inscrip-

tion, although known as a Syriac name, V=^= , and as Hebrew,
(373 Num. 13") and Nabatean (But. 213").

II. This tessera is circular.

Obverse : fV

The name HfV should perhaps be read VlDrV, for there
seem to be traces of a yod. This name, as I believe, occurs
here for the first time. For *|"TT see No. 1.

Reverse. This shows the portico of a Greek temple. On
either side is a palm-branch. The door is double, each leaf hav-
ing two panels, the upper trellised, the lower decorated with an
outline like a Latin fl. Some traces of characters in the gable
may be read as 'HN

my brother

In this case the tessera would commemorate the death of the
brother of Yathbelit. However, the reading may also be

TIN-
III. This tessera is sexangular.

398 H, H. Spoer, [1906.

Obverse. A figure, reclining upon a funerary couch, dressed
in tunic and mantle, the head covered with the modius. The
bare right arm is stretched out, to receive a cup offered by a boy,
who in the other hand (left) bears a pitcher. Above, a medal-
lion, the bust of a man, in a beaded frame. (Cf. with tessera
described by the author in this Journal, xxvi, p. 114.) The
inscription is hardly legible :

* * * •pvjjf

Reverse:

Three busts in a row, a fourth above, in the middle flanked
on either side by a sun in splendor. Between the busts are
small bosses.

IV. This tessera is a rectangular oblong, one side only hav-
ing been used, the reverse being rough and shapeless.

Obverse. Between a row of bosses, above and below, the
word 'n^D

Belli

A name which occurs also in I. To the left a rayed star.
'Reverse. None.

V. Small, square and of a terra cotta color.

Obverse. A reclining figure on a funerary couch. Below :

YedVbel

A name which is known ; cf . Lidzbarski, Ephemeris I, p. 203,
206.

Reverse. A winged female genius with a trailing garment
holding a chaplet in her extended left hand. On the left mar-
gin, '"10 , known as a Nabatean proper name.

VI. Square. Figures as described by me in this Journal,
xxvi, p. 115, 111. 5.

Obverse. ?V(1

Halwan

The form of the last letter is unusual (but see Lidzbarski,
Schrifttafel II, Rom. 1). The name is known as Nabatean

i ° "
n.pr.m,, ,jU^&. . The third letter is evidently not "1, so that

we have not here the well known name
Reverse. Below the couch :

The last letter may possibly be }<•

Vol. xxvii.] Notes on some Palmyrene Tesserae. 399

Three letters, barely legible, occur near the left margin:

VII. A rectangular oblong inscribed on both sides. There
are no figures.

Obverse. Much worn :

The third letter is partly erased. Unless the deity be referred
to (Vog. No. 138), the name Sems does not elsewhere occur
alone, but only as compounded with other names.

Reverse. [*J] ^

Beni Bezbol

This is one of the many names compounded with 'TO » but .1
have not Elsewhere met with this combination.

VIII. Of the form of an olive.

Obverse. In the center a bust, perhaps of a deity; below it
a large boss, and above an ill-preserved inscription which I read

Jlel-Sedeq

A name which occurs for the first time. Cf. the Hebrew
piy, Gen. 1418.

Reverse. The field is occupied by a horse, saddled, the sad-
dle supporting a rod crowned by a five-pointed star. Before
the horse is a boss, and beneath, a curious symbol which may
be described as a crutch-head enclosed in an arch. May this be
the fragment of the Swastika, which occurs not unfrequently in
association with Apollo?1 The star symbolises Istar= Venus.
The horse is sacred to the sun-god Sems, 2 Kgs. 13". That this
belief was held by the inhabitants of Palmyra seems to be borne
out by another tessera in my possession, which bears on the
one side the sun-god and upon the other a horsed chariot with its
driver. Unfortunately the head of the man is worn away, and
the tessera somewhat mutilated.

1 Cf. Thomas Wilson : The Swastika, p. 852.

Three Objects in the Collection of Mr. Herbert Clarlt, of Jeru-
salem.— By GEORGE A. BARTON, Professor in Bryn Mawr
College:

No. 1.

This little weight, in bronze, in the form of a turtle, came, 'Mr.
Clark said, from Sebastiyeh. The drawing is about its actual
size. It is inscribed with old Hebrew letters : £'£11 ,• evidently
standing here for the "fifth" of a shekel. The weight is 2^
grams, or 58 grains. At this rate the shekel of which it was a
fraction consisted of 290 grains. A series of half -shekel weights
found at Tell Zakariyeh weighed 157.5 gr., 146.7 gr. respec-
tively (Bliss and Macalister, Excavations in Palestine 1898-1900,

146 ff., and Clermont-Ganneau, Recueil d'arch. orient., 25 ff.),
and one in my possession weighs 153.5 gr. (cf. JAOS. xxiv,
386 ff.). The limits of variation of the Hebrew shekel have
never been determined, but it is evident that there was a shekel
which approximated 300 grains. Mr. Clark's turtle weight is
one-fifth of that. The turtle form of this weight is, so far as I
know, unique.

No. 2.

This object, cut out of a blackish stone, bears on one side a
face; on the other, pictures of two birds accompanied by cer-

tain marks. The face, which the drawings exhibit in profile as
well as in full view, has a physiognomy which resembles the
faces on two Hittite monuments published by Messerschmidt,

Vol. xxvii.] Barton, Three Objects. 401

Corpus Inscriptionum Hetiticarum, II, Tafel xxxiv, A and B.
The resemblance to the profile of A is particularly striking.
One of the bird figures resembles a figure on Tafel xlvi of the
JVachtrag of Messerschmidt's work. I am of the opinion,
therefore, that the object may be Hittite, and put forth the sug-
gestion in hope that some one who has given more attention to
the decipherment of Hittite may be able to throw further light
on it. The object may have been used either as a weight or a
seal.

No. 3.

This little stone duck, found at Sebastiyeh, was perhaps a
weight, though this is by no means certain. It weighs 39 grains.
If it were intended for a weight, it was probably an eighth of a

shekel, since eight times its weight would be 312 grs. Since
the shekel varied so much, however, it may have been the
seventh of a shekel, since 273 grs. is almost equal to 277.8 grs.
— the smallest of the shekels found by Bliss at Tell Zakariyeh.

VOL. xxvii. 27

Studies of Sanskrit Words. — By EDWIN W. FAY, Professor
in the University of Texas, Austin, Texas.

I. Arvanc-

IN his German etymological dictionary (English translation,
s.v. Erde) Kluge, with some hesitation, connects Lat. arvum
'field, ploughed land' with Gr. c/aa£e 'earthwards.' In his Eng-
lish Etymology, s.v. earth, all this hesitation disappears, in
view, I take it, of ON. jorvi 'ground,' not cited in the former
work among the cognates. It is also to be noted that Vergil
(Aen. 12,681) uses arvis as a virtual rendering of Gr. xaP°%>* 'to
the ground.'

There is a very obvious difficulty in the comparison of arvum
with «?/oa£e, viz., the vocalism. This difficulty may be sur-
mounted, to the best of my knowledge, in but one way, by
assuming a base er- 'arare,' to which the European base ar-
(from ar-) would belong. This solution I have offered in my
essay entitled A Semantic Study of the Indo-Iranian Nasal
Verbs (Am. Jr. Phil., 26,389). There is, it must be admitted,
not very much to justify a base er-. Skr. ara 'awl': Gr. d/us
' auger ' is dubious because of OHG. ala. Latin ora ' edge,
point,' is also ambiguous, and so is er ' stachel-schwein,' which
may well be for ' her', if Hesychian XVP *8 genuine Greek,
with a genuine x- Not. very conclusive is Skr. irinam (? from
ar-) " brilchiges, unfruchtbaresland." A vague connection with
the root might be made out for Gr. £/n<£os ' kid ' (if = " pricket ") :
Lat. aries 'ram' (cf. Umbr. erietu) : Lith. eras (with e) 'lamb.'
But in any case, the vowel relations of arvum and €pa£e are no
harder to admit than those of Lat. aries : Umbr. erietu.

As a means of testing the correlation of arvum with €pa£c I
have made a study of the Sanskrit word arvaiic-, as it appears
in the Rig-Veda, and I attach some importance to the isolation
of its formation, with the adverbial suffix -anc-, as compared
with the isolated Greek word l/aa^e. I conclude from the data
that follow that arvaiic- and the adverb arvak do mean "earth-
ward(s)."

Vol. xxvii.] Studies of Sanskrit Words. 403

1. The Petersburg Lexica define our word as follows: a)
herwaerts kommend, hergekehrt, zugekehrt, entgegen-kom-
mend: b) diesseitig (only in grammatical and lexicographical
sources, save the adverb arvak, which has literary attestation) :
c) unterhalb befindlich, nach unten gerichtet: d) in der nahe
von (advb. only). The earliest citations for the sense c) are
from AY. and SB., and consist of passages in which arvanc-
is contrasted with urdhvas. This contrast I would, however,
see in RV. vii. 78. Ibc (B,)1:

urdhva asya anjayo vi srayante,

uso arvaca brhata rathena — vamam asmabhyam vaksi,
" On high her rays are spreading abroad,
<^Now^> Usas earthward in thy mighty car wealth to us
fetch."

This hymn may belong to a late middle period of RV. tradi-
tion as Arnold believes (see the table in PAOS. xviii. 353,
and KZ. xxxiv. 341), but the chronology of the Vedic hymns
can hardly tell against a word proved to be proethnic ; and a
hymn of the earliest period, if liable to popular rather than
hieratic iise, may have retained in all its working-over very
early linguistic material (cf. Bloomfield, Proceedings, xxi. 41 ff.
pace Arnold, ibid. xxii. 309 ff.).

2. Contexts are not wanting in RY. where a special sense
seems warranted for arvanc-. Thus in i. 92. 16 (A),

— vartir asmad2 a ... arvag ratham . ni yachatam
"unto our house— earthward — your chariot bring",

arvag might well be taken as a mere adverbial repetition of
vartis, and passages are still to be pointed out in which arvak
seems to mean "ad nostrum fundum." In some of these con-
texts previous translators have recognized the sense I would give
to arvanc-, and Sayana glosses arvak in v. 45. 10 (see 5 below)
by avanmukhah 'face-downwards.'

1 After each hymn I put Arnold's indication of its age: A, archaic:
JBi, early middle period : B2, late middle period : Ci, early late period :
C2, later late period.

2 This takes asmad as a poss. adj. = 'nostrum', identical with the
compounding stem asmad.

404 K W. Fay, [1906.

viii. 14. 8 (Ba) :

ud ga ajad angirobhya avis krnvan guha satfh

arvancarii nunude valam,

"Showing the hidden, he drove forth the cows for the

Angirases,

And Vala he cast headlong down " (so Griffith) .
That arvaiicam nunude means ' struck to the ground ' = (' to his
feet,' cf. RV. i. 32. 8, B2) seems to me most probable. Note
the combination of the root nud with urdhvam ' upwards ' in i.
85. 10 (B2);i. 88. 4 (A).

Wilson, cited by Griffith, paraphrased x. 71. 9a (C,),

ime ye narvan na ^aras1 caranti,
by "Those who do not walk (with the Brahmans) in this

lower world, nor (with the gods) in the upper world.'

3. In the following passages arvanc- is combined with the
root sad 'to sit.'

iii. 4. 8cd (B2) : sarasvati sarasvatebhir arvak

. . . barhir edam sadantu,

"May Sarasvati and her confluent rivers earthward

(come) . . . and sit down upon this grass."
x. 15. 4a (CJ : barhisadah pitara uty arvag,
' ' Grass-sitting Manes come earthward to our aid (or

with aid)."

4. In the following the contrast of earth and heaven is clear,
i. 45.10ab (A) : arvaiicam daivyam janam agne yaksva,
"Fetch-by-sacrifice to earth the celestial kind, O Agni."
v. 83. 6cd (C3) arvan etena stanayitntinehy

apo nisincann asurah pita nah.
" Earthward in this thunder come,
dripping water, god our father."

vi. 19. 9cd (B2) : a visvato abhi sam etv arvan
indra dyumnam svarvad dhehy asme,
' ' From every side let him come earthward :
Indra, bestow upon us heavenly glory."

1 On paras, see 6 below.

Vol. xxvii.] Studies of Sanskrit Words. 405

vii. 83. 3 (B2) : sam bhumya anta dhvasira adrksata

indravaruna diyi gh6sa aruhat

asthur jananam upa mam aratayo

arvag avasa havanasruta gatam.

' ' The earth-ends were seen dust-bewhirled ;

Indra-Varuna, my cry mounted to heaven :

My enemies among the peoples encompassed me :

Earthward with help, hearing my cry, ye came."

x. 83. 6a-c (Cj) : upa mehy arvan — manyo vajrinn-

" Come earthward to me, lightning wielder, Manyu."

5. In the large majority of instances arvanc- is of vague
significance. The gods are called upon to come and bless the
worshipper. Obviously ' ' come hither " makes a satisfying ren-
dering for arvan yahi, but "come down, come earthward"
were equally satisfying. Arguing from the antiquity of the agri-
cultural rites of worship, attested for instance at Rome in the
ritual of the Fratres Arvales, and mindful of the etymological
sense of ' ploughland ' found in arvum, one might feel disposed
to make something of

iv. 57. Gab (C,) : arvaci subhage bhava site vandamahe

tva,
"Earthward, gracious one, turn thou, Furrow, we greet

thee."

Still, as so many other deities are summoned earthward, we can
hardly lay much weight on the summoning of the ' ' Furrow "
earthward.

The following passages, though vague, have seemed, for one
reason or another,1 worthy of consideration.

ii. 37. Sab (B2) : arvancam adya yayyam nrvahanam

ratham yunjatham iha vam vimocanam.

' ' To come earthward to-day your man-bearing

Car hitch up; here is your unhitching."

Here arvancam is not attributive, but predicative, a sort of
factitive to yunjatham; unless construed closely with yayyam.

1 Chiefly because they show arvanc- in combination, not directly with
a verb of motion, but with one equal, by a sort of zeugma, to a verb of
motion.

406 E. W. Fay, [1906.

iv. 4. 8a (B2) : arcami te sumatim ghosy arvak,
"I sing thy grace; sound it <back> to earth."
v. 45. lOcd (BJ : udna na navam anayanta dhira
asrnvatir apo arvag1 atisthan,

' ' Like a ship in water the wise launched him (the sun) ;
The hearkening floods <[of light descended^* to earth
and stood <there>.

vii. 18. 3c (B2) : arvaci te pathya raya etu,
"Earthward (? ad fundum nostrum) come thy path of

wealth, <^Indra>.

vii. 28. Ib (A) : arvancas te harayah santu yuktah,
"Earthward be thy steeds yoked, <^Indra^>."

viii. 61 (50). 1 (A): ubhayam srriavac ca na indro arvag

idarh vacah

satracya maghava somapitaye dhiya savistha a gamat.
"Let Indra hearken earthward to this our double song;
<^Moved^> by our unanimous prayer, let Maghavan, the

mighty, come hither to drink soma."

x. 89. 5d (B2) : narvag indram pratimanani debhuh,
"Nor have any counterfeits decoyed Indra to earth."
x. 89. 16d (B2) : tiro visvan arcato yahy arvan,
"Past all (other) praisers (?), come down to earth"

(ad fundum nostrum: cf. vii. 18. 3, and paragraph
2, above).

x. 129. 6c (C2) : arvag deva asya visarjanena.
"The gods are later than this world's production"

(Griffith).

No translation of this passage is likely to win conviction, even
from its proposer, but I venture on

"<^There came^> gods to earth at its creation."

6. I have reserved two passages for separate treatment,
i. 164. 19ab (C,) : ye arvancas tan u paraca ahur
ye parancas tan u arvaca ahur,

1 Note the gloss of Sayana, mentioned above at the beginning of 2.

Vol. xxvii.] Studies of Sanskrit Words. 407

"The down <on the earth> they call the up <in the

sky>;
The up <^in the sky> they call the down <[on the

earth>."

In this stanza we have some astronomical or cosmogonic riddle,
and a real solution I do not pretend to offer,1 but it is probable
that paranc- here repeats para- in stanzas 17 and 18, where para-2
connotes 'heavenly,' as in the same stanzas avara-, echoed in
stanza 19 by arvanc-, connotes ' earthly.' This interpretation
shows points of agreement with Wilson's paraphrase of x. 71. 9a
(see in 2 above).

viii. 8. 23 (A) : trfiii padany asvinor avih santi guha

parah
kavi rtasya patmabhir arvag jivebhyas pari,

' ' The three regions of the Asvins are revealed
<^ which were^> in hiding in-the-f ar-heaven ;

The two seers of righteousness wing-their-way earthward
unto the living."

The translation of parah by " in-the-f ar-heaven " ( — German
"jenseits"), rather than by "before," seems to me beyond
question; and this would seem to fasten the sense of "earth-
ward " upon arvak.

After the above tests of the special applicability of the ren-
dering "earthward," it seems not amiss to regard Skr. arvanc-
as a cognate of Gr. e/oa£e, Lat. arvis (in Aeneid 12. 681).

Postscript.

The editors have asked me to add, for completeness' sake, a
word on (1) arvacina- (arvacina-) and (2) arvavat-. It is habit-
ual to render (1) — construed like Lat. sublimis (Gildersleeve-
Lodge's Gram. §325. 6) — by 'hitherward' (='to the worship-
per, to me'), rather than by 'earthward,' and in all the usage
of the word (fifteen cases) there is nothing, as Professor Hopkins
observes, to prove the inadequacy of the usual rendering. The

1 But we may note the Vergilian usage, Aen. 6. 481, of super! = ' qui
in terra (supra terrain) sunt,' for the usual super! = ' caelestes.'
* See Grassmann's Lexicon, s.vv. para-, parama-.

408 E. W. Fay, [1906.

one difficult use is RV. vi. 25. 3 (A, in Arnold's system of
dates) :

indra jamaya uta ye 'jamayo

'rvacinaso vanuso yuyujre

tvam esaih vithura savansi

jahi vrsnyani krnuhi paracah,

as to which Grassmann remarks in his Lexicon that here alone
arvacina- is used of other than friendly approach. But there need
be no question of approach at all, for we may well take jamayah
and ajamayah as adjectives and arvacmasah as a substantive, in
formation something like arvales, but in sense like vicini (fini-
tumi) or Landsleute. The stanza does not lose in point thereby:

Indra, our kindred and non-kindred

Neighbours, that as enemies have united, —

Do thou in sunder their mightiness

Rive, their prowess; make them as strangers (=drive
them afar).

In the study of arvavat ' proximity,' the salient fact is that
it is never used save as an antonym of paravat ' distance ;' but,
after a consideration of all the examples in RV., I feel free to
say that ' distance ' is not the only signification of paravat (and
its kin). Grassmann's Lexicon s.vv. para- parama-, suggests
' heaven ' as a rendering, and renders parastat by ' oberhalb ' ;
while both Grassmann and the larger Petersburg Lexicon inter-
pret tisras paravatah by ' the three regions ' (sky, air, and earth) .

The connotation of ' sky ' or ' air ' (' aloft, on high ') seems to
me probable — what connotation is mathematically demonstrable ?
— for this group in the following instances. In iv. 26. 6 (B2)
the falcon brings the soma-stalk from the paravat (a-b), having
taken it divo amusmad uttarat 'from yon remote sky' (d),
and padas ab, without the interpretative clause d, recur in sub-
stance in x. 144. 4 (A). In iv. 21. 3 (A), Indra is besought to
come from (1) divah 'the sky,' (2) prthivyah 'the land,' (3)
samudrad . . purisat 'the sea-of-air, — i. e., from the three
regions already mentioned : the stanza then adds (4) svarnarat
' from the light-realm ' and (5) paravato va sadanad rtasya ; I
interpret (4) as a substantial repetition of (1) and in (5) I take
sadanad rtasya, which Sayana glosses by meghalokat ' from the
cloud-space,' as a synonymous (explanatory) apposition with

Vol. xxvii.] Studies of Sanskrit Words. 409

paravatah (which Ludwig renders by an adjective) : thus (5) =
'from paravat, the cloud-space.' In yi. 8. 4 (B2) Matarisvan
is said to have brought Agni down from (his hiding place in)
paravat, while in x. 187. 5 (B2) Agni's birthplace is given as
pare rajasah ' in the far-off of the air.' In v. 53. 8 the Asvins
are summoned from the sky (divah), the air (antariksat) , and
from here (amat), and besought not to remain afar, paravatah
'from (=in) the paravat. Further, note viii. 12. 17 (A):

yad va sakra paravati samudre adhi mandase

asmakam it sute rana sam indubhih,
"Whether, O Might, thou joyest in paravat in the sea

<of air> [So Griffith supplies]
Delight in our pressing," etc.

In addition to these examples of the connotation ' sky ' (air)
for paravat (param), we may note the contrasting pair avara-
lower (and) parama- ' highest,' especially in i. 164. 17 (C,) avah
parena para ena 'varena'. In the light of such instances we
may note that in the remote Celtic branch Ir. eross, which Stokes
(Fick's Woert.4 II. p. 37) gives as a cognate of Skr. para-, means
' height,' which would tend to vindicate the sense of ' high ' for
proethnic pero-.

If paravat means ' sky,' what of its antonym arvavat ? Note
viii. 13. 15 (A) :

yac chakrasi paravati yad arvavati vrtrahan

yad va samudre andhaso 'vited asi,
"Whether, O Might, thou art in paravat, or in arvavat,

Vrtra- slayer,
Or else in the sea <of air>, thou art the protector of the

Soma-stalk."

If we are right in taking samudre of the 'air,' then paravati
and arvavati are the sky and earth, respectively.1

1 [Observe, however, that the same words are repeated in viii. 97 (86).
4, but filled out (after the invitation is given) in 5 as follows: yad vasi
rocane divah samudrasya' dhi vis^api, yat parthive sadane vrtrahantama
yad antariksa a gahi, " or if thou art in the sky's brightness, (or) on the
sea's expanse, (or) if on earth's seat, (or) if in the interspace (air), do
thou come hither," where ' sea ' is distinct from air, and earth and sky
are separately contrasted. ED.]

410 K W. Fay, [1906.

The semantic problem may be stated as follows : para- meant
(1) ' distant, far ' but came, by a connotation which may have
been proethnic, to mean (2) 'high, in the sky;' its antonym,
arvanc-, meant (1) ' earthwards, towards (on) the ground ' but
developed, under the influence of para (1), the meaning (2)
' near.'

The following illustration furnishes an approximate parallel.
In Latin, domi (domum) and apud me (ad me) became, in a
restricted sense, synonymous: ' at (to) my house.' These syno-
nyms must have played a role in the upgrowth of domo doctus
for a me doctus and of domi habeo aliquid for mihi est aliquid
(cf. Lorenz ad Mil. Glor. 194), wherein the sense of ' domus '
has nearly vanished. Similar is the generalisation of dvpa^t ' out, '
French fors / hors ('Lat. foris), from which the sense of 'door'
has vanished, almost or wholly ; and in French chez the sense of
Lat. casa is all but gone ; and we no longer think of a hill .when
we say down or adown. In general, on such prepositional
words (direction adverbs) derived from nouns, consult Steinthal-
Misteli, Abriss der Sprachwissenschaft,* II, §4, p. 11 ff., noting
especially Skr. parsvam parsve- 'adversus, ad, apud, prope.'

To say briefly what I think of the morphology of the group
under discussion, I explain arvanc as a terminal accusative
*arvam (or plur. *arvan)-(- a deictic particle *-c(a), comparable
with Gr. -8e; *-c(a) may be compared with Lat. -ce, and if it-
belongs to a different guttural series, the reason is that *arvams-
has been attracted into a group with the other direction adverbs
in -anc-. Alongside of arvacma RV. exhibits a pretty large
group of which pracma and praticina may be taken as representa-
tives ; arvavat is not to be explained as from arvanc, but merely
as a counterpart of paravat.

2. Nahus-.

In RV. viii. 8. 23 (above), the words trlni padani call for in-
terpretation : what are the three padas ? Sayana interpreted them
as the three wheels of the Asvins' chariot. Griffith says heaven,
firmament, earth. But the hymn itself mentions three places
from which the Asvins come, viz., nahus- (stz. 3), antariksa-
(3, 4), dyaus (4, 7). Dyaus we know and antariksa- we seem
to know, but what is nahus? I believe nahus to be 'the night,'

Vol. xxvii.] Studies of Sanskrit Words. 411

cf. Gr. vvxa ' vvKTtap, Ivwxos. This interpretation yields good
results when applied to

vii. 6. 5bcd (B.2) : yo aryapatnir usasas cakara

sa nirudhya nahuso. yahvo agnir

visas cakre balihrtah sahobhih,

"Agni made the dawns noble-spoused,

Driving off the nights, strong Agni

Made the peoples tribute-bringers by his might."
Here note the opposition of usasas and nahusas.

The base to which I ascribe Gr. vv^a and nahus is s)no(w)-gh-,
alternating with s)ne(y)-gh-, and refer for my conception of
the phonetic problems involved to Am. Jr. Phil. xxv. 371 ff.
379 ff. Stripped of "root-determinatives," the base in simpler
form is s)ne(y)-/s)no(w)-, and meant "to wrap," cf. Skr. sna-
yati 'wraps,' Lat. nuit glossed by 'operuit, texit.' The word
nahus belongs more closely with nahyati 'binds, wraps,' while
Gr. vvxa has tae vowel color of nuit. Lat. niger ' black ' and
noegeum ' amiculi genus ' attest the -ey- diphthong. In all this
it has been assumed that the night was the "binder" or "wrap-
per up" of the day (cf. Am. Jr. Phil. xxv. 386, note 2). The
base for "snow," with a different final guttural, s)ne(y)-gwh-,
has a cognate meaning, 'snow being conceivable as that which
"wraps" (covers) the earth: cf. also Avest. vafra-: 'snow':
the root vap- "to strew, weave."

3. vedhas, 'worshipper, pious; faithful, true.'

Uhlenbeck in his etymological lexicon groups together vida-
tham (with deaspiration) 'congregation, assembly,' vidhati
'worships, honors, dedicates (to a god),' and vedhas as defined
above. For none of these words does he suggest further cog-
nates, not even Avestan ones.

In view of the uncertainty in some few Sanskrit words, even
the oldest (cf. Whitney, Verb Hoots, sub the root vrh, and
Wackernagel, Altind. Gram., § 161), between b and v, we may
provisionally etymologize on our words as though they began
with b. Then if we set down *bedhas- 'fidus, pius,' it becomes
immediately apparent that *bedhas and fidus are etymological
cognates, which differ only in their stems, the former being an
-6S-/-OS- stem, the latter an -e-/-o- stem. However, it must be
observed that in AV. *xix. 3. 4 the stem vedha is found in a

412 E. W. Fay, [1906.

variant reading for v6dya, while in old Latin fidusta (from
*fidos-to-) occurs, defined by Paulus as "a fide denominata, ea
quae maximae fidei erant," a definition that would lead us to
infer an Italic stem *fidos- : cf . also f oedus and conf oedusti.

The derivation of vedhas here suggested also accounts for
vidatham, if etymologically defined by "federation." But
vidhati presents a harder problem. It would not be very well
defined by irelOu., but is fairly well matched by Germ, betet,
beten and bitten, being, according to many,1 cognate with 7r«'0a,
fidit. But if Kern is right in referring these German words to Skr.
badhate 'premit' (cf. the citation of the footnote), then it might
be necessary provisionally to separate vidhati from vedhas, and
rather put it in a group with badhate. I have tried, however, in
Am. Jr. Phil. xxvi. 179 ff., to reunite 7rei0« and badhate under
the still remoter base bhe(y)-d(h) 'to split > <^splice.' The
semantic questions involved will now justify, I hope, a some-
what more detailed treatment, apropos of the problem presented
by vedhas for bedhas.

Beside the root bheidh ' to convince ' — though this meaning is
far from being primitive — stands a root bheid 'to split.' My
thesis is that these roots were originally but one. The variation
of aspirate and sonant at the end of roots with nasal infixes is
far too common a phenomenon to be called in question, and the
nasal inflexion of Lat. findit ' splits,' Skr. bhinatti, bhindanti"
fulfils the conditions. Further, a nasal inflexion of bheidh
seems attested by Alb. bint 'I persuade,' if they are right who
connect it with Gr. W0ei.3

If thus on the formal side we may regard bheidh-/ bheid as
one root, it remains to bring in accord the figurative meaning of
bheidh ' to convince ' and the direct sense ' to split ' attested for

1 E. g., Osthoff, cited in Uhlenbeck's got. Woert, s. v. bidjan ; Brug-
mann, Orundr. i. § 589; Kluge, Etym. Woert. s.vv.; Skeat's Concise
Etym. Diet., s.v. bid.

* It is entirely within the bounds of probability that bheid- ' to split '
has been affected by a group-association with the root of Lat. scindit,
Skr. chinatti, chindanti.

3 So, among others, Brugmann, in his Grundriss, and Prellwitz, Woert. ,
s.v. neiOu. G. Meyer, Alban. Woert., s.v. bint, derives from bhendh,
which is, in my opinion, itself but a derivative of bhe(j)dh-; see Am.
Jr. Phil. xxvi. 181.

Vol. xxvii.] Studies of Sanskrit Words. 413

bheid-.1 If we define Lat. fidus by 'open(ed), frank, sincere,
loyal' we see how it may be cognate with findit 'splits, opens.'
Similarly we may define the Homeric idiom Tretdetv <£p«/as (dvfiov)
TIVI (TIVOS) by 'to open one's mind, convince,' comparing our own
idiom "to open one's eyes;'2 fidit and WTTOI^C may be rendered
by 'opens (one's own heart) to (another), trusts.' The concrete
sense of ' splits ' is perhaps retained in Iliad 15, 26 7r«rt0crt5o-a
0veAAas, which may be defined by ' findens procellas.' The locu-
tion jiuo-flaJ Trf.i6f.iv ' to bribe ' is very like the Sanskrit compound
dana-bhinnas ' bestochen, bribed.'

To Lat. f6edus I assign a semantic development somewhat
different from that exhibited by fidus 'true.' In Sanskrit the
ptc. bhinnas means not only ' split, opened,' but is defi'ned in
the lexicon of Boehtlingk by "verbiinden mit, hangend-, haft-
end an;" cf. vi-bhinnas " unzertrennlich verbunden mit," sam-
bhinatti " zusammen bringt, ' in Beriihrung bringt, verbindet,
vermengt, sich zu jemandem gesellt," bhiduras "in nahe Beriih-
rung tretend — sich vermengend, sich vermischend mit," bhittis
"a woven mat" (: Lat. fides 'strings, a lyre'). With these
words Lat. foedus 'truce, league, compact' accords in defini-
tion and they show that in Sanskrit derivatives, at least, the
root bhid- has developed the connotation 'to join.' Similar is
Eng. splices 'joins (split rope-ends)', whose derivation from
splits is clear; also, in the language of weavers and rope-makers,
Germ, scheren 'to cut, shear' has acquired the sense of ' spannen;'
cf. also Eng. pieces 'to join <^pieces]> together.' The seman-
tic opposition of "to split" and "to join" is only apparent, and
comparable with the conflict found in the pair sticks ' stecht '
and sticks 'steckt' (cf. also stitches 'stecht, stickt;' and see
Kluge's Woert. s.v. stechen).

But in demonstrating a root bheid(h) 'to split,' with the con-
notation 'to join,' the last word has not yet been said for foedus
'truce.' Touching foedus, I think of some primitive form of
contract by indenture, some breaking of a tessera hospitalis, in
which the breaking of the token was the chief symbolic act of

1 1 share Uhlenbeck's doubts whether Goth, beidan can be directly
connected with Lat. fidit 'trusts.' But in view of MHG. stecken 'to
remain fast, stick, bide ' we may connect beidan, Eng. bide directly
with bheid(h) to split, pierce.'

2 Cf. Fr. resoudre ' to persuade,' from Lat. resolvere ' to open up' (?).

414 E. W. Fay, [1906.

the treaty-making. Thus do we best account for the idiom
opKia. TTHTTO. Tafielv ' f oedus ferire, icere, percutere,' which lends
itself to the interpretation " symbola <^pactionis> fissif acere "
rather than to ' ' f oederis causa < hostianT> sacruficare" No
doubt, however, the cutting up of the animal sacrificed for
distribution among the compact-makers was a part of the cere-
monial (cf. Aristophanes, Lysis., 192;? Vergil, Aen. 8. 641).

Returning now to vedhas 'fidus,' I conjecture that its ortho-
graphy with v for b was primarily due to the association of forms
of b(h)eidh 'to split, pierce' (cf. Goth, beidan, Eng. bide, Gr.
TTtinOova-a ' as explained above) with forms of the Sanskrit root
vyadh vidh ' to pierce ' (cf . Lat. di-vidit) ; or, to put it con-
cretely, I conjecture that *bedhas ' apertus, aperto <^pectore^>,
' fidus ' has been assimilated to viddhas ' di-visus, apertus ;'
though it is of course not to be denied that vedhas, defined by
'apertus, etc.' is susceptible to immediate derivation from vidh-
yati 'peirces.'

4. vadhri.

The close kinship of vadhri and its Greek synonym Wpis
' To/xuas , castratus ' is not to be called in question, despite their
phonetic divergence. The phonetic difficulty is resolved by
deriving vadhri from the Sk. root vadh 'to beat, slay,' and ifyus
from a base widh-, found in Skr. vidhyati ' pierces,' Lat. di-vidit
' divides," and further attested, I believe, in to-fyids (from *widh-
tmos, with -tmos from the root tern), ' (mare) dividens.' The
parallelism of vadhris (*wedhris) and Wpis (widhris)' throws light
on the Skr. root vyadh (not attested in RV.), which I take to
be a blend of the root vidh (with grade forms in vedh) and
the root vadh.

5. sprsati 'touches, grasps, feels, besprinkles.'

•

Uhlenbeck (ai. Woert. s.v.) finds no sure cognates for sprsati,
but mentions the possibility that Goth, faurhts ' fearful ' (with-
out s-) is identical with the ptc. sprsta ' touched, stirred, moved.'
[Professor Hopkins calls my attention to the fact that sprsta
appears as prsta in RV. i. 98. 2.] I doubt not, however, that

1 On the close correspondence of Latin and Sanskrit in parts of their
vocabulary, see Kretschmer, Einleitung, 125. ff.

Vol. xxvii.] Studies of Sanskrit Words. 415

Gr. o-Trapao-o-ei ' tears, rends, mangles ; mulcat' is cognate with
sprsati 'touches; imilcet;' cf. Lat. tangit 'mulcet, mulcat.'
Goth, faurhts leans to the violent sense of 'mulcat;' as a seman-
tic parallel we may compare Gr. eKTrAayets <'terror-> smitten.'
Wharton (Etyma Latina) sets down spurcus as a cognate of
sprsati, and I believe this to be correct. We may again illus-
trate by derivati ves of tangit, viz., contaminat, contingit 'defiles,'
contactus 'defiled; cf. also tangit 'smears.' The German word
f erch ' dung ' (without s-) perhaps belongs more closely with
spurcus. With these we might group Lat. porcus, supposing
the pig to have been named (1) ' the dirty creature,' and not (2)
' the rooter ' (porca ' furrow ') . Still another possibility for
porcus is (3) ' spotted, dark,' cf. Skr. prsni ' spotted, a cow,'
Gr. Trep/cos, TrcpKvds ('spotted), dark,' irepK-rj 'perch' (a dark or
spotted fish, cf. our fish-name of "spot"). There is no incon-
sistency between (1) and (3), ' dirty ' and ' spotted ' being closely
related notions, as Lat. maculosus, e. g., shows. And if porcus
meant ' rooter ' (2) , it may still be a cognate of virapdwu ' tears,
rends.' I see no reason to doubt, either, that prsan-f, defined by
Boehtlingk as " sich anschmiegend, zartlich " (mulcens) belongs
with sprsati (cf . also Whitney, Roots, Verb-Forms, etc, etc. , sub
prs), cf. upa-sprsati "zartlich berilhrt, liebkost." With this
group we may classify the cognates of Lat. procus ' suitor,'
precatur 'entreats,1 presses (a suit, request); lacessit, flagitat.'

6. khudati ' futuit.'

Uhlenbeck defines khudati by "stosst hinein (kaprtham,
sapam), and (s)khidati by " reisst, stosst, driickt." No cog-
nate of khudati seems to have been pointed out.2 If the long
diphthong gradation -e(y)-/-6(w), already referred to in this
paper, is correctly assumed, then khudati and khidati go back to
a common root (see Am. Jr. Phil. xxvi. 396). So far as signi-
fication goes, khudati would seem but a specialization of khidati,
and we might explain its vocalism as something individual, due,
to use the metaphor introduced.by Bloomfield (IF. iv. 78), to a
blend of khidati and its synonym tudati ' stosst, sticht, stachelt.'

1 Eng. entreats derives from Lat. tractat ' handles ; ' cf . further, Goth,
bidjan: Skr. biidhate 'premit' (supra, p. 412).

J But now cf . Prellwitz, Woert. 2 s.v. Kvadoq ; I would derive nva6oc from
khudhtos, Lat. cunnus from khudhnos or khudnos : base khud(h).

416 E. W. Fay, [1906.

But the infection of khidati by tudati may as well have begun
in the primitive period as in the separate life of Sanskrit. Uhlen-
beck remarks s.v. khidati, " verwantschaf t mit chinatti is nicht
undenkbar." In Latin both (per-)scindere (= chinatti) and per-
tundere occur in the special sense of khudati,1 the former in Priap.
15. 5, 54, 77. 13, and the latter in Catullus 32. 11. If scindit
and tundit thus cross meanings in Latin, we have some confirma-
tion of the supposed association of ideas that changed khidati
to khudati under the influence of tudati.

If Lat. cudit ' strikes, beats ' corresponds with Skr. khudati
' stosst hinein,' the recognition of the Italic cognate would for-
bid us to regard khudati as khidati inffected by tudati. It
would not forbid us to suspect that primitive khudeti is khideti,
with the vowel color of tudeti, though we should be bound to
admit three roots meaning 'to strike, thrust, pierce, split,'
whose weakest forms are; 1) (s)khid2, 2 (s)khud, 3) (s)tud, (cf.
Uhlenbeck, op. cit., s. v., tomaras). The derivation of (2) from
(1), inflected in its vowel color by (3), is purely glottogonic;
not in any case a phonetic question, but rather a psychological
question. Provisionally, leaving out the possibility that khudati is
cognate with Lat. cudit ( : Germ, hauen, cf . Brugmann, Grund.
I8 §639), we may include khudati / khidati among cases like
those pointed out by Bloomfield in the essay referred to. Ulti-
mately, perhaps, a psychological treatment of the vowel alter-
nation in the spirit of Wundt's Die Sprache (I1, p. 335 ff.)
may be arrived at.

7. Skr. ambaram.

Uhlenbeck asserts that no satisfactory explanation has been
advanced for ambaram, but it seems to me that an easy one lies
at hand. The meanings we have to account for are (1) ambitus,
vicinia, (2) amictus. It is phonetically allowable to connect
amb- with ap.<f>i, Lat. ambi-, cf . Skr. ambu / ambhas ' water '
for the variation b / bh after nasals. By this explanation
ambaram (subst.) is morphologically comparable with avara-

1 Cf. Gr. icpovei, which occurs in the same special sense.

2 Strong form (s)kheyd ( : Lith. skedziu, Lat. caedit, see Hirt, Ablaut.
67) : cf. Amphitruo, 159, quasi incudem caedant, where incudem caedant
partakes of the nature of the etymological figure, as does the commoner
locution incudem tundere.

Vol. xxvii.] Studies of Sanskrit Words. 417

'lower,' apara- 'further' (adj.): ambara- means 'the sur-
rounding.' For the sense ' amictus,' note the "roundabout" of
the sailor, and the style of cloak called "circular." Compare
too Sanskrit vasah paridhanam.

8. Pratlcih in RV. iv. 3. 2d (BJ.

Agni is summoned in this stanza to a fire kindling. The
general purport is clear, but there is a verbal difficulty in the
last pada,

ima u te svapaka pratlcfh,

to wit, as to what substantive is to be supplied with praticih.
Sayana supplies ' flames ' or ' hymns,' Ludwig ' gentes ' or ' cives '
or even the ' ladles of the sacrifice,' and Grassmann supplies
' libations,' while Griffith follows Sayana. In support of the
native interpretation I cite vii. 39. Ib (A)
pratici jurnir devatatim eti,
"The toward flame goes to the godhead."

Here the situation is that the fire has been kindled and the flames
ascend. In iv. 3. 2, Agni is invited to come and kindle the fire
and the ' toward <fflames^>' are the ' expectant flames ' unless,
instead of jurnayah, we supply samidhah 'kindlings, faggots.'
As to svapaka, Ludwig's ' selbst garer ' suggests to me ' self-
cooker, self-kindler.'

VOL. xxvii. 28

Notes on the Mrcchcikatika. — By Dr. ARTHUR W. RYDER, The
University of California, Berkeley, Cal.

THE following notes serve as a supplement to my translation1
of the Mrcchakatika, recently published as Volume ix of the
Harvard Oriental Series.

The method of citation here adopted is a slight modification
of that explained on pages xv-xvi of the edition and translation
of the Karpuramanjari, Volume iv of the Harvard Oriental
Series. The verses are cited by the act and the number of the
individual verse within the act. The citation for prose gives
the number of the act, the number of the last preceding verse,
and the number of the prose speech counted from the last pre-
ceding verse. The following abbreviations require explanation :
Apte. Practical Sanskrit-English Dictionary. By V. S. Apte.

Poona, 1890.
Bohtlingk. Mrkkhakatika . . . tibersetzt von Otto Bohtlingk.

St. Petersburg, 1877.
Calcutta commentary. The commentary of Sriramamayasar-

man, found in the Calcutta edition of 1870. This comment-
ary I have been obliged to take at second hand.
Godabole. The Mrichchhakatika . . . edited by N. B. Goda-

bole. Bombay, 1896. Bombay Sanskrit Series, No. lii.
HOS. Harvard Oriental Series.
JV. Jivananda Vidyasagara. Mi'ichhakatika (sic) . . . edited

with a full commentary by Pandit Jibananda Vidyasagara,

B.A. Third edition. Calcutta, 1898.
LD. The commentary of Lalladiksita, as given in Godabole's

edition.

Levi. Le Theatre Indien, par Sylvain Levi. Paris, 1890.
P. The commentary of Prthvldhara, as given in Parab's

edition.

1 The Little Clay Cart . . . translated ... by Arthur William Ryder.
Harvard Oriental Series, Volume Nine. Cambridge, Massachusetts,
1905.

Vol. xxvii.] Notes on the Mrcchakatika. 419

Parab. The Mrichchhakatika . . . edited by Kashinath Pan-
durang Parab. Bombay, 1900.

Regnaud. Le Chariot de Terre Cuite . . . traduit . . . par
Paul Regnaud. Paris, 1876.

Stenzler. Mrcchakatika . . . sanskrite edidit A. F. Stenzler.
Bonn, 1847.

Wilson. Select Specimens of the Theatre of the Hindus, trans-
lated . . . by H. H. Wilson. Volume i, pp. 1-182. Third
edition. London, 1871.

It is greatly to be regretted that we have no old and full
commentary on the Mrcchakatika. The only comment which
may be earlier than the nineteenth century is the meager gloss
of Prthvidhara, who gives us no information about his date ; it
is interesting to note (see Parab, 37. 25) that he knew the
Brhatkatha. But Prthvidhara builds upon previous commenta-
tors; he refers to a pracinatika (e. g. 40. 28), to ke cit (e. g. 7.
25), to eke and apare (e. g. 26. 25-26). Unless this previous
exegesis should come to light, we must remain in doubt 'about
many points, especially in the Prakrit of Samsthanaka and the
Candalas. Yet, even so, the Mrcchakatika is one of the easier
of the works of the classical Sanskrit literature. Sudraka's vocab-
ulary is not very large, his sentence-structure is simple, and his
thought is rarely involved or difficult. Inasmuch as the action
of the play continues for only five or six days,1 the author does
not use the Viskambha or the Pravesaka.

In HOS. ix, p. xix, I have called attention to the fact that
Sudraka does not slavishly follow the canons of dramaturgy as
laid down in the technical works which we possess. It is worth
while in this place to give examples proving that his grammar
also conforms less closely to the norm than that of Bhavabhuti,
for example.

(a) In i. 14, under stress of meter, he uses nidhanata in the
sense of nirdhanata. JV. warns us that we must pardon the

1 See HOS. ix, pp. xxvi-xxix. There is nothing to show whether the
action of the tenth act occurs on the same day as that of the ninth act.
Windisch, Berichte der philol.-histor. Classe der Konigl. Sachs. Oesell-
schaft der Wissenschaften 1886, pp. 474-479, allows only four days for
the action of the play. But the speech of Viraka, ix. 23, shows that a
night had passed between the strangling of Vasantasena (Act viii) and
the trial (Act ix).

420 A. W. Ryder, [1906.

blemish, because Sudraka is a great poet: nidhanasabdo marane
rudhah, atra tu dhanasunyatve prayuktatvat prasiddhivirodha-
khyadosah sodhavyah, mahakavipranitatvad iti bodhyam. Cf.
Hit. i. 128 (134), where this verse reappears with variants but
with nidhanata. In i. 37, nirdhanata is used.

(b) In i. 32, he uses the causative form namyati. This does
not prevent his use of the normal namayati in prose, at viii. 46 *.
Compare the causative form unnamya, used by Bhavabhuti in
Malatimadhava, ix. 31.

(c) iii. 18s. The masculine singular desakalah is curious;
we should expect either desakalau or desakalam.

(d) v. 30. Here it is hard to parse yadvat, though the
meaning is plain enough. But this case is complicated by the
reading of the second line; see the note on this verse, below.

(e) vii. 4. sa tavad asmad vyasanarnavotthitam. If this
reading is correct, we have the ablative asmat agreeing with
the prior member of the compound. Compare, in vii. 8, the
curious use of asmin without a noun : so etasmin in Ratnavali,
ii. 19.

(f) viii. 38. The 'use of asraya as a feminine is, so far as I
can find, elsewhere unparalleled.

(g) x. 27. The phrase dusitam yasah is logically coordi-
nate with maranat, and should therefore be in the ablative.

Such instances might be multiplied. In the structure of his
verse, also, Sudraka permits himself some liberties. Instances
in point are i. 30, where the third pada is in a different meter
from the rest; iii. 7 (fourth pada); iv. 17 (second pada). Levi
.has shown (pp. 206-208) in a masterly fashion that such con-
siderations have little value for the dating of the play ; but they
are of real importance in the exegesis of cases less certain than
those given.

Act i.

i. 1. It is of negative interest to observe that Sudraka's
Nandl invokes the favor of Siva. Of course, this is the case
with the great majority of the plays whose authors are without
sectarian bias, and regard their works as pieces of literature
without didactic purpose. The sequence of thought in the
first three lines is rigid. Line 1 suggests the physical means
employed by Siva to plunge himself into trance ; line 2, the con-

Vol. xxvii.] Notes on the Mrcchakatika. 421

k

sequent numbing of the organs of sense; line 3, the resultant
insight into things as they are.

In line 4, the long compound sunyeksanaghatitalayabrahma-
lagnah is not wholly simple. The commentators and transla-
tors differ widely in their interpretations. It seems to me that
the Calcutta commentary offers the best explanation : sunyam,
prapancabhavah : tasya yad iksanarii darsanam tena ghatito jato
yo layas cittaikagrata tatpravanataviseso va, etc. The line
may then be literally translated: "May the meditation of Sam-
bhu protect you, which is fixed on the supreme being with an
intensity sprung from his insight into the emptiness of the
material world."

i. 4. The information here given about the life of Sudraka
is tantalizingly imperfect. P. tells us that the phrase agnim
pravistah means that he made a sacrifice of his body in the fire,
as did the old sage Sarabhaiiga: see Ramayana (Bombay ed.)
iii. 5. 38, pravivesa hutasanam or Raghuvansa, xiii. 45, ciraya
samtarpya samidbhir agnim yo mantraputam tanum apy ahausit.
Similar phrases are used in the Mudraraksasa, where Visnudasa is
reported to be burning himself alive from grief at the loss of his
friend Candanadasa: vi. 15" jalanam pavesidukamo ; vi.1512 agni-
pravese; vi. 16s hutabhuji pravesahetuh ; vi. 17" jalanam pa vi-
sami.

i. 7. See note on iii. 30, below.

i. 8l. For the expression annamaarii jialoam pekkhami com-
pare Nagananda iv. 131 : putta aharii kkhu tujjha maranabhida
savvam pi jialoam garudamaam pekkhami.

i. 82J. This speech is interesting in showing how the mean-
ing ' cut ' may have come to attach itself to the causative of
kip. Kappijjantam plainly signifies here 'cut to pieces, killed,'
in its application to Jurnavrddha ; but it also means ' arranged '
as applied to the young bride's hair. Both meanings are
included in the English ' fix '. The secondary meaning of ' fix '
in such slang phrases as "I'll fix him," is quite like the second-
ary meaning of kalpayati. This explanation seems more natu-
ral than that of the PW.1 The word is used in the sense of
' cut' again in i. 30 and in iii. 213'4.

1 [For a parallel in the development of meaning, compare sud (arrange,
' fix,' kill). ED.]

422 A. W. Ryder, [1906.

i. 12l. The phrase dasie putta atthakallavatta has been, I
think, quite misinterpreted by the translators. Wilson has:
' ' The sons of slaves ! your guest is ever ready to make a morn-
ing meal of a fortune." Regnaud: " Ah les fils d'esclaves! Us
font un dejeuner de votre bien." Bohtlingk: " Diese Sohne
von Sclavinnen mit ihrem Bischen Gelde." These translations
assume that arthakalyavarta is a bahuvrihi : ' whose breakfast is
money ; ' but the same word in ix. 221 must mean ' a trifle of
money ' : compare strikalyavarta in its Prakrit form in iv. 52, ' a
mere trifle, namely a woman;' at ii. 124>B>131 the word kalya-
varta also means a 'trifle.' These translations also miss the
point of khajjanti; it is not supposed guests, but the money
itself, which makes itself at home only where it isn't used for
food (khajjanti), like the cattle-boys who stay only where they
are not eaten up (khajjanti) by wasps. The correct interpreta-
tion is found in JV., who says: dasyah putra ity anena cai
'sam (i. e. arthanam) atiheyata sucita. In other words, dasyah
putrah is merely a humorous epithet of arthakalyavarta : ' this
damned money-trifle.' The phrase dasyah putrah is in Prakrit
not infrequently applied to things, with the same illogical
humor found in the corresponding use of ' confounded ' or
'damned' in English. Compare also "You son-of-a-gun of a
fool' (confounded fool). In iii. 6!4 it is applied to the gem-
casket; in v. 471, to the storm. In Sak. vi. 20l (Pischel) and
in Nagananda iii. 25, it is applied to bees. It will be noticed
that in all these cases the phrase is used by the Vidusaka.

i. 14. This verse reappears as Hitopadesa i. 134 (Godabole
and Parab) with the following variants: line 1, sattvat pari-
bhrasyate for prabhrasyate tejasah ; line 2, nihsattvah for niste-
jah; line 3, sokanihatah f or sokapihitah.

i. 15. The phrase vairam aparam means 'another (form of)
hostility (with mankind) ' ; the figure is striking. In the second
line svajanajana- must mean 'kinsmen and strangers,' as the
commentators say. The ca in the third line is difficult: P. and
LD. say cakaro hetau; Bohtlingk and Regnaud adopt this sug-
gestion. But this meaning for ca can hardly be found else-
where; is it not better to understand it in the sense of 'if (cet)?
Then the line means: " (it is the part of) wisdom to go into the
forest if (ca) there is (bhavati) contempt from his wife."

i. 22. Sudraka uses visesayati in the meaning of ' surpass '
again at iv. 4.

Vol. xxvii.] Notes on the MrcchakatiJca. 423

i. 23. It cannot be doubted that Samsthanaka's arithmetic is
at fault here ; the ' ten names ' are really eleven, and are not to
be reduced by combination ; dase 'ti vyartham, says P. tersely.

i. 305. With the expression na puspamosam arhaty udyana-
lata, compare iv. 6 : no musnamy abalarh. vibhusanavatim phul-
lam iva 'ham latam.

i. 309. The phrase savami bhavassa sisam attanakehirii pade-
him is repeated almost exactly at viii. 3718. In spite of the
differences in case, it can hardly be doubted that Samsthanaka
means 'I swear by the gentleman's head and by my own feet,'
not ' . . . schwore ich . . . mit meinen Ftissen beim Haupte
dieses klugen Herrn,' as Bohtlingk translates.

i. 31. The last line is repeated at viii. 17'.

i. 32. For the form namyati see above, page 420. [It may
be due to confusion with -am roots of the ya- class (tamyati,
etc.). The middle (passive) namyate also suggests the form.
-Ec.]

i. 323. One of the subtlest points in the character of the
ignorant and conceited Samsthanaka is the fact that he permits
the Vita to apply to him repeatedly the drastic epithet kaneli-
matr, apparently never realizing the gross insult.

i. 34. Quoted in Dandin's Kavyadarsa, ii. 226, 362.

i. 344. It seems as if the grammar would be better if bhusa-
nasabdam and malyagandham were in the nominative.

i. 41. The initial esa 'si harks back, with humorous effect,
to the esa 'si of the preceding verse.

i. 45. Parab's nirmitah is apparently a mere misprint for
nirjitah.

i. 46. For the form suskavan, see Pan. viii. 2. 51.

i. 503. The word kakapadaslsamastaka, which Samsthanaka
applies to Maitreya again at ix. 30" receives two explanations
from the commentators. It is said to mean either (1) chief of
the princes of sharpers, or (2) whose head-pate is like a caret.
The first explanation seems forced and unnatural, the second
quite in keeping with Samsthanaka's character; slsamastaka
iti sakaravanitvena punaruktatvam na dosah (P.).

i. 5019. This speech (repeated at i. 553> B) contains, so far as I
know, the only reference to a female stage-manager (sutradharl).
For a word like tandavasutradhari (comm. nartane kusala) in
Caurapancasika 7 is obviously without bearing on the history of

424 A. W. Ryder, [1906.

the drama in India. The expression in our passage cannot be
used as direct evidence, since it may be nothing more than
another absurdity of Samsthanaka's.

i. 51. The words line a vele have received a double San-
skritization from the commentators; either as linayam sa vela-
yam, or as rnam ca vairam. The sense which Bohtlingk and
Regnaud extract from the second interpretation is excellent:
"a pumpkin stalk . . . , debts, and enmity never rot"; but the
construction is very strange, with a string of nominatives fol-
lowed by na khalu bhavati putih. Unless we are willing to
take the ungrammatical and illogical form of statement as a
part of Samsthanaka's dialect, we are forced to adopt the first
explanation: "a pumpkin-stalk ... do not rot, even when a
long time has passed" (JV. : velayam samaye linayam atitayam
api).

i. 52. What does nirvalkalam mean as applied to a sword?
It seems to me that the word is used in an extravagant sense ;
the sword is barkless, i. e., it has no time to gather mould, it is
always busy. JV. suggests that it means ' out of its sheath '
(valkalam tarutvak, laksanaya tannirmitaih kosam, tasman nir-
gatam: niskosam ity arthah). This seems quite out of accord
with kosasuptam in the next line ; but perhaps, after all, Sam-
sthanaka would speak of a ' naked ' sword as one ' ungarmented,'
and if so, the immediate contradiction of kosasuptam would be
but another absurdity of the speaker. Samsthanaka seems
interested in radishes; cf. viii. 34.

i. 526. Bohtlingk supplies rohasenam as object of pravrnoti;
but. there is no indication of Rohasena's actual presence, nor
is there any reason why she should do this apavaritakena.
Surely, we have to supply atmanam, as in the common stage-
direction apavarya: she wraps herself in the mantle, without
letting Carudatta see. Indeed, Bohtlingk supplies atmanam
with pravrnoti below at ii. 207.

i. 56. The phrase bhagyakrtam dasam probably refers to
Carudatta's poverty, not to Vasantasena's profession, as Boht-
lingk takes it. Carudatta deplores, almost too frequently, his
plight, but he does not often refer to the fact that Vasantasena
is a courtezan. Besides, this fact surely would not prevent her
from entering the house. The two concluding lines are diffi-
cult. They seem to mean: "and because of her acquaintance

Vol. xxvii.] Notes on the Mrcchakatika. 425

with men, she does not speak impudently, even though she
speaks many things." But Vasantasena has not spoken a word
aloud since her entry into the house ; so that perhaps we have
to understand ' a man ' as subject of bhasate. This seems to be
Wilson's understanding of the verse ; he renders :

"Nor makes she harsh reply, but silent leaves
The man she scorns, to waste his idle words."

i. 56s. The sense seems to require at the end the phrase
utthedha tti which is found in Stenzler and Godabole, but is
lacking in Parab and JV.

i. 57. The striking expression timiranikara, 'multitudinous
darkness,' is used again by Bhavabhuti at Mai. viii. 1. Nikara,
'heap,' corresponds exactly to the negro use of heap, in 'heap
dark,' etc.

Act ii.

ii. 0'. The verb alikh is explained by LD. as meaning 'to
yearn ' (abhilas) ; Regnaud and Bohtlingk follow this explana-
tion, the former reading karii pi, the latter, kim pi.. Perhaps
the verb, in combination with hrdayena, does acquire this mean-
ing ; but it seems more natural to take it in its ordinary meaning
of ' draw, paint': " she is painting something (or ' somebody,' if
we read karii pi) with her whole heart." The something is, of
course, a picture of Carudatta, perhaps the one which she is
gazing at in'iv. O1. A lovelorn heroine in Indian drama has no
moi'e familiar occupation than painting the picture of her
beloved.

ii. O14. The text is doubtful; my translation simply attempts
to make the best of Parab's text, but the result is not wholly
satisfactory. The reading which JV. offers deserves considera-
tion : ko kkhu nama ajja attabhodie anugahido mahusave taru-
najano — What young person now receives my mistress' favor in
the great festival (of her joy; mahan utsava anandasvarupas
tasmin) ?

ii. 2'. We may read devi bhavissarii as two words 'I will
turn goddess,' or devibhavissarii as one word, 'I will turn god.'
The situation is reproduced in Kipling's Krishna Mulvaney.

ii. 4". Parab's reading kridati seems inferior to krldatah of
Stenzler and Godabole, and seems to have no Ms. authority.

426 A. W. Ryder, [1906.

ii. 6S&. The chaya should read dhurtye instead of dhurtay-
iimi.

ii. 9. In Parab's text, tretahrtasarvasvah should be printed
as one word.

ii. 10l. In this speech and in ii. 11s the word tapasvi seems
to be used in a double sense ; on the one hand it means ' poor,
unfortunate (varaka) ' and on the other hand, ' saint. ' The
second meaning is made almost certain by the fact that each of
these speeches is followed by a verse describing ascetic prac-
tices. Then ayarii tapasvi, like ayam janah, will refer to the
speaker, not to Mathura, as LD. and the translators understand,
and the little speech will mean : ' ' and yet, what more should a
poor saint like me do? for I, etc." This is the view which JV.
presents: tapasvi varako nirdoso 'ksama iti va: ayam ity atma-
nirdeso dardurakah kim karisyati karotv ity arthah.

ii. 13" (just before ii. 14). The word tulidam is difficult;
probably it means 'proportioned (to her innocence),' as the
same word in ii. 14 means 'proportioned (to his strength).'

ii. 141. Godabole's suggestion that lakkhida mhi represents
in Sanskrit raksito 'smi is worthy of consideration.

ii. 14". Read kam for the kim of Parab's text.

ii. 1426 (just before ii. 15). The phrase kudo so dhanio is
intentionally ambiguous. On the one hand, it means: " Where
is your creditor (i. e. Mathura)?" On the other hand, it means:
"How can he (Carudatta) be a rich. man?" The shampooer is
shrewd enough to see that it is the hidden meaning which really
interests Vasantasena, and answers accordingly.

ii. 15. Both the meter and the interpretation offer difficul-
ties. On the meter, see Stenzler, pp. 257-258. The meaning
of lines 3 and 4 depends on the presence or absence of na, given
in Parab's text and assumed as present by P., but having very
little Ms. support. If na be kept, then we must, with P.,
assume that the last line contains a rhetorical question ; and so I
have translated. But. I confess that the latter half of the verse
puzzles me greatly.

ii. 162. The reading nam is preferable to Stenzler's na, and
has more Ms. authority.

ii. 167. Here Parab's reading seems inferior to the ajjo
bandhuanam samassasidum of Stenzler and Godabole.

ii. 168. The phrase is one of exaggerated courtesy: "mis-
tress, if it may be, then let this art remain in the hands of a

Vol. xxvii.j Notes on the Mrcchakatika. 421

servant (of yours)," that is to say, "take me into your service."
Regnaud translates: "... permettez-moi d'exercer mon art
a votre service," which renders the idea freely; but Bohtlingk
seems to me to miss the point with his translation : " . . . ge-
statte, dass ich diese meine Kunst den Handen deiner Diener-
schaft anvertraue." JV. is very clear: parijanahastagata pari-
janasya posyajanasya mame 'ti bhavah . . . sevakatvena mam
anumanyasve 'ti bhavah.

ii. 17. Doubtless Stenzler is right in printing bihaccham
(bibhatsam). It is hard to see how vlhattham could represent
a Sanskrit vihastam, and the efforts of the commentators to
explain the latter word are far from satisfactory.

ii. 18. The edi (eti) of the other texts seems preferable to
Parab's ehi.

ii. 191. Parab's tac ca seems inferior to the tarii ca of Stenz-
ler and Godabole.

ii. 19s. The phrase vamacalanen-a judalekkhaam ugghusia
ugghusia has caused commentators and translators considerable
difficulty. It is plain from the word dyutalekhaka that the
monk is none other than the shampooer; this much P. has seen.
The word lekhaka is used in ii. 2, where it plainly has the
meaning 'a (gambling) score.' Stenzler's chaya is right, I
think, in taking ugghusia as the representative of udghrsya
rather than of udghusya. The phrase then becomes simple
enough: "stumbling with my left foot over a gambler's score."
The suddenly metamorphosized shampooer has forgotten to rid
himself of his gambling paraphernalia, which he drops when
attacked by the elephant.

ii. 206. The custom of marking a garment with the owner's
name is referred to again in viii. 43n.

Act iii.

iii. 2. For Parab's annapasattakalatte we have another read-
ing annakalattapasatte. Either gives a good sense.

iii. 3. LD. suggests that priyatama virahaturanam may be
taken as one word : ' of those sick because of separation from
the beloved.'

iii. 81. The svapitah of Stenzler and Godabole is better than
the svapiti of Parab and JV.

iii. 12. P. and LD. explain darsanantaragata as meaning

428 A. W. Ryder, [1906.

'found in the treatises on robbery,' and this, I think, is cor-
rect: cf. darsitah in iii. 12'. Regnaud also follows the com-
mentators on this point.

iii. 13. The commentators take vistirnam as a separate form
of breach, and so obtain seven forms in all, according to the
passage from the Cauradarsana which they quote. This is a
point on which it is best to follow their authority, but Boht-
lingk (p. 190) states the objections. '

iii. 14. Regnaud and Bohtlingk take visamasu in the sense of
' difficiilt ' ; but I think LD. is right in giving it the meaning
'unsuccessful' (viparitasu). The epithet then anticipates and
explains the dosan of the fourth line.

iii. 16J. The words cikitsam krtva are better taken as part
of the text, not of the stage-direction.

iii. 17'. Parab's na in the phrase kva na khalu salilam bha-
visyati is apparently a mere misprint for nu.

iii. 181. It seems to me that the translators miss the point in
tan mama 'pi nama sarvilakasya bhumistham dravyam when
they interpret : ' ' whatever is underground is my property "
(Wilson). Does not the speaker rather mean :"well, the prop-
erty belonging to me too, to Sarvilaka, is underground," that
is, "I have no property?" If this is correct, the clause is
merely a humorous afterthought to the preceding sentence.

iii. 183. Apparently Parab's kim na is a misprint for kim nu.

iii. 1.8 B. On desakalah, see above, p. 420. The construction
of dhikkrtam andhakaram (bis) is quite unclear to me; the
translators take dhik krtam as two words : ' fie on the made
darkness.' This construction seems very forced and awkward,
but I can offer nothing better. I think, however, that a mark
of punctuation should precede bhadrapithena : ' fie on the dark-
ness caused by the bhadrapitha ; or rather, fie on the darkness
caused by me,' etc. The reading asmadbrahmanakule of Stenz-
ler and Godabole seems better than the -kulena of Parab and
JV.

iii. 19. The word anirveditapaurusam is very ciirious; we
should expect anivedita-, 'to which manliness is unknown,' and
this is what JV. reads. Perhaps we have to parallel this form
with nidhanata for nirdhanata in i. 14 (see above, p. 419),
but the meter does not demand the form anirvedita- here.

iii. 20. Read in Parab's text vag desa-.

Vol. xxvii.] Notes on the Mrcchakatika. 429

iii. 21". It is strange that Maitreya should quote the words
dudiam via duaraam ugghadidam, when Radanika has not used
the expression in what precedes.

iii. 24. This verse is repeated as v. 43.

iii. 265. -The sentence beginning bhaavam kaanta is repeated
almost literally at vi. O80, this latter time in Vasantasena's mouth.

iii. 29. This verse is repeated as v. 7.

iii. 29 l. With asmaccharlrasprstika (a gesture of solemn
asseveration) cf. the expression marjaro bhumim sprstva karnau
sprsati in the fable of the cat and the vulture in the first book
of the Hitopadesa. JV. has: gatrasamsparsena sapathakara-
nam laukikaprasiddham eva.

iii. 30. In the second half of this verse the words na yasya
raksam are difficult of interpretation. This difficulty doubtless
explains the presence of the variant nrpasya raksan, which
Stenzler adopts. This latter reading gives a good sense, and
forms the basis of the translations of Regnaud and Bohtlingk.
But the large majority of the Mss. have na yasya raksam; and
as this is the lectio difficilior, we are bound to make what we
can of it. It is easy to see how na yasya raksam might be
altered by a puzzled reader into nrpasya raksan, while the
reverse process is almost inconceivable. The commentators give
little help. P. and LD. are silent. JV. has: yasya sandhe rak-
sam na pariharami na tyajami satatam eva sandhim raksami 'ty
arthah. This reads almost like nonsense, for it makes the second
half of the verse contradict the first half; the last thing
which Carudatta desires is the preservation of the breach. The
Calcutta commentary is hardly more successful; yasya sandhe
raksam raksanam svarupena 'vasthanam iti yavat : na pariharami
no 'pekse sandhim raksitum na saknomi 'ty arthah. Here the
explanation of raksam as ' ' the preservation (of the breach) in
its present form " is ingenious enough ; but the commentator is
forced to give to na pariharami a meaning exactly the reverse
of that which the phrase should have. Surely na (sandheh)
raksam pariharami must needs mean : " I do not avoid the pre-
servation (of the breach) in its present form," which is precisely
the opposite of what the speaker wishes to say. Wilson trans-
lates as if the na were absent: "we'll leave no trace to catch
the idle censure of men's tongues." The translation of Wilson
and the brave attempt of the Calcutta commentary point the

430 A. W. Ryder, [1906.

way, I think; to the solution of the difficulty. Obviously, the
trouble lies in the na. The clause becomes plain enough if we
read nayasya, which involves no change in the Ms. reading.
The verse may then be translated: "Quickly close up the
breach with these bricks; I avoid the preservation of justice,
because of the abundant evil of scandal." The word naya
occurs once more in the play, in i. 7: nayapracaram vyavahara-
dustatam. In this passage the words naya and vyavahara are
used, it seems to me, with a slight double entendre. The line
means, on the one hand, "the practice of legal justice, and the
vicious quality of a legal process," as illustrated in the ninth
act of the play, and, on the other hand, "the practice of justice
(referring to Carudatta) and viciousness of conduct (referring
to Samsthanaka)." This affords a further suggestion for iii. 30.
Carudatta means to say: "I am willing, under the circum-
stances, to thwart the law," and perhaps he wishes his auditors
to, understand nothing more than this; but to himself he means
to say: "Scandal is such a dreadful evil that I am justified in
departing from the course of conduct which strict justice
demands." The word naya, with its more specific and its
more general meaning, like "justice" in English, is admirably
adapted to express both the artha and the bhava, the super-
ficial and the deeper meaning.

Act iv.

iv. 1. We should surely read raksan instead of raksyan.

iv. 3. The second line is explained by iii. 12; it was unlucky
for a thief to see a woman during his expeditions. JV. remarks
that the caurasastra forbids a thief to enter such a house as
that here described.

iv. 71. Delete the mark of punctuation after alamkarao in
Parab and Godabole.

iv. 7". Stenzler is right in printing the iti as part of the
stage-direction.

iv. 14. The slight illogicality in the singular -sumanah is
doubtless owing simply to the stress of meter, and is not worth
the trouble which the commentators give themselves to ex-
plain it.

iv. 17. The meter is irregular (12. 11: 12.12).

Vol. xxvii.] Notes on the MrcchaJcatika. 431

iv. 20. In the third line, va has more authority than ca, and
is probably the correct reading; it seems to be used in the
sense of eva (JV. has va avadharane).

iv. 241. JV. takes rastriya to mean chief -of -police (rastra-
palah: nagararaksayam niyutko rajapurusa ity arthah), rather
than brother-in-law of the king; and I think he is right, for
Samsthanaka nowhere appears as a government officer, giving
sensible orders in Sanskrit. On the other hand, rastriya is
used at ix. 384 and x. 514 in the meaning 'brother-in-law of the
king.' As Bohtlingk remarks (p. 192), we should have an iti
at the end of the speech.

iv. 252'3. This IK the only indication in the play that Sarvilaka
is the son of Rebhila.

iv. 274. There seems to be a pun on puspaka, and I have
translated accordingly. The reading naaranari is better than
Parab's naranari.

iv. 276. The pompous language of this description of the
portal makes one wonder whether it is not an intentional trav-
esty. In the long compound beginning with torana-, JV.
explains -vedia- as 'pedestals' (talasthabaddhapradesah).

iv. 2710. There is doubt about the form and meaning of
kuraccuatellamissam. If the second element represents San-
skrit -cyuta-, perhaps it means ' drippings.' P. has bhaktatai-
laghrtamisrapindam ; but JV. analyses quite differently: kurad
dravyavisesac cyutam nisthyutam yat tailam tena misram yuk-
tam. The wordkura is used again in x. 29.

iv. 2712. If sahma- really belongs here (it is lacking in many
Mss.), it probably means 'own'; the whole word will then
mean 'possessed of its own dice, made out of gems.' Read
paribbhamanti in Parab.

iv. 2714. Here pagldao must be used in an active sense,
'singing' (prakarsena ganapara ity arthah: JV.). Sasingarao
(sasrngarah) is impossible ; we must read either sasingarao (fern,
plu.) or sasingaraam.

iv. 285. On the tame madana^arika, see HOS. iv, page 229,
note 8. To pesianti LD. supplies yoddhum, 'are provoked to
fight.' In Parab's text, pandikida is a misprint for pindlkida.

iv. 29. This passage (ma dava . . . loassa) is printed by
Stenzler, Godabole, and JV. as prose ; only Parab regards it as
a verse. The matter is of a sort which the Vidiisaka would be

432 A. W. Ryder, [1906.

apt to put into verse ; it is not narration nor description, but a
kind of humorous moralizing. If we regard the passage as
verse, it falls into five padas, the scheme of which is 12, 17: 13,
14, 15. Now if we disregard the fifth pada, and apply the
rule padantastham vikalpena (Srutabodha 2) or va padante
(Vrttaratnakara, i. 9), we obtain the scheme 12, 18: 12, 14,
which makes a pretty fair arya ; and it seems to me that Parab
is right in assuming that the Vidusaka speaks here in verse.
But what becomes of the words anahigamanio loassa ? It is of
course possible that they were intended by Sudraka as a prose
remark following the verse; biit this is improbable. The words
add nothing to the sense of the passage ; they are merely an
explanation of the comparison of Vasantasena's brother with a
graveyard champak. Is it not probable that these words were
originally a gloss? It would be natural for a reader to add the
marginal comment anabhigamaniyo lokasya ; the next scribe
might easily incorporate the remark into the text, and the easy
change into Prakrit would naturally follow. We have an inter-
esting parallel in vi. 20s. Here Candanaka, after his Prakrit
speech, adds in Sanskrit : kim sabdavicarah : strlpunnapumsaka-
vyakhyanam aprastutam. "Why consider the words? An
exposition of .feminine, masculine, and neuter is irrelevant."
Although all the Mss. give this matter, it is rendered very sus-
picious by the unmotivated change into Sanskrit, and has all
the appearance of being a glosa. Parab omits it; and JV.,
omitting it in the text, gives it as a part of his comment. I
think there can hardly be a doubt that JV. has correctly pre-
served the original division between text and comment; and if
this is true, it seems probable that the same process has taken
place in iv. 29, with the added step of changing the isolated
Sanskrit words into Prakrit.

iv. 29 '. The commentators take phullapavaraa- to mean ' a
garment embroidered with flowers ; ' and in view of the variant
pupphapavaraa-, this is perhaps to be preferred to the meaning
' expanded, baggy.'

iv. 29". The correct reading of the word which Parab gives
in the form kavatthadainie is doubtful; and P.'s interpretation
of kavattha- as = kapardaka does not help. We can hardly do
better than read karatta- and adopt the explanation 'dirty.'

iv. 30". Read in Parab' s text accharia- (ascarya-) for accha-
ridi-, and write rovida aneapadava as two words.

Vol. xxvii.] Notes on the Mrcchakatika. 433

iv. 32 '. As soon as the conversation becomes familiar, Va-
santasena reverts to Prakrit.

Act v.

v. 2. The compound in the first line is thus analyzed by
LD. and JV. : jalardramahisasyo 'daram bhrngas ca tadvan
nilah.

v. 5. To the word patrachedya (cf . chedya ' engraved ') the
commentators unite in giving the conventional (rudha) mean-
ing 'picture,' and we can hardly doubt that they are right, as
patra is used in the same way with other words implying mark-
ing upon a (leaf) plate.

v. 6. In the second line, va = iva. In the third line adhva-
nam means ' road ' in reference to Yudhisthira, but ' silence '
(a-dhvana) in reference tQ the kokila.

v. 7. This verse is the same as iii. 29.

v. 71. LD. has an artificial explanation of the last clause,
according to which dustah = dosah, and the whole is to be taken
as ironical. The translators seem to follow his lead. Boht-
lingk, however, points out (p. 196) that dustah should be neu-
ter, if this explanation is right. But a simple, literal transla-
tion gives better sense and better Immor : ' ' there even rogues
are not born," i. e. a courtezan, an elephant, etc., make it
impossible for anything, even a rascal, to flourish. This is also
JV.'s understanding of the passage: dustah sadosa api jana na
jayante na tisthanti 'ty arthah: dosatirekasya 'vasyambhavad iti
bhavah.

v. 91 In kamo vamo, the Vidusaka makes use of his third
homeiy prose proverb in this scene. This is one of the touches
which make Maitreya a living character, very different from the
stock Vidusaka.

v. 93. Stenzler and JV. are wrong in rendering avedha by
apeta (= apagacchata). Of course, it represents aveta, 'under-
stand.'

v. II2'5. This little scene is imitated by Harsa in the Ratna-
vali, ii. 68 \

v. 11*. We would welcome an a (ca) after andhaare. But
compare duddinandhaare in v. 38".

v. II14. The verb kakaasi shows pretty plainly that indama-
hakamuko here means 'a crow' (so P. and JV.), not 'a dog.'
VOL. xxvu. 29

434 A. W. Ryder, [1906.

v. II18. Parab's punctuation, with the mark after tena hi, is
suggestive. The expression seems very colloquial: "first guess
it, man ; then (you may do as you like) . "

v. 11B1 (just befoi-e v. 12). In Parab's text, suvassabhandaam
is, of course, a misprint for suvanna-.

v. 15. The word nirantarapayodharayfi is used in a double
sense. In reference to the night as a natural object, it means
' whose clouds are close together ' ; in reference to the night as
a rival wife, it means ' whose breasts are close together (i. e.
swelling).'

v. 152. In my translation, I have taken strlsvabhavadurvi-
dagdhayfi to mean 'ignorant of woman's nature,' because this
meaning seemed to fit the context better than ' obstinate because
of her woman's nature ' ; but the latter meaning, I now think,
seems more natural to the word, and is probably correct.

v. 18. The commentators are doubtless right in taking pro-
sitabhartr as a feminine, 'whose husbands are distant,' though
in prose we should expect prositabhartrka.

v. 19. In Parab's text, balakapandurosnlsam should be
printed as one word.

v. 20. JV. explains protsarya by apasiirya : ' the clouds have
driven away and captured the moonlight.'

v. 30. Parab's reading nirapeksa seems to me better than
niraveksya, especially as I find no other instance of the com-
pound niraveks. The construction is elliptical, but easily
intelligible: "as (was thy grief when) thou didst speak falsely
. . ., such is my grief also; O cruel! Let the cloud be
restrained."

v. 36. In the fourth line, Parab's reading -mukha- seems to
me much better than the -sukha- of the other editions. The
genitives then modify -mukha-, as if we had daksinyapanyasya
mukhasya niskrayasiddhir astu. The verse may be literally
translated : ' May you have success in the sale of your face, the
birthplace of fraud, deceit, and lies, together with pride; con-
sisting of perfidy, in which love-sports have made their home ;
the courtezan's stock-in-trade, the compendium of amorous fes-
tivals; the price of which is courtesy.'

v. 40. The use of adita eva is unusual ; it is precisely equiva-
lent to our colloquial 'from the start,' 'from the word go.'

v.,42. The commentators are sorely troubled by drstapurva-
samgamavismrtanam, and offer very forced and artificial explan-

Vol. xxvii.] Notes on the Mrcchakatika. 435

ations. Does it not mean simply ' forgotten in the gatherings
of their former associates ' ?

v. 421. The bath-clout is that mentioned in iii. 18s.

v. 43. This verse is the same as iii. 24.

v. 50. The compound pracalitavedisamcayantam is difficult,
and the commentators not wholly satisfactory. Probably vedi
means ' pedestal ' (cf . note on iv. 27") ; samcaya (aggregation)
perhaps means ' a construction of closely -joined bricks (LD.
militabhir istikabhir nirmanam)': then vedisamcaya will mean
'the brick- work of the pillar-pedestals.' The whole compound
will therefore signify ' by which the edges of the brick- work of
the pillar-pedestals are shaken.' In other words, the awning,
flapping in the wind and rain, threatens to tear out by the roots
the pillars to which it is fastened. In the fourth line, Parab's
text should read samklinna for saklinna; the word means
* soggy.'

Act vi.

vi. O30. The words bhaavam ... purisabhaadheehim are
repeated almost literally from iii. 26 6.

vi. O46. The chaya should read tvarate for tvarayati; the
latter would be in Prakrit tuvaredi.

vi. 1. It is possible to understand the long compound in the
first and second lines in either of two ways: " the great ocean of
misery and woe, called (apadesa = vyapadesa) the king's
prison ' ; or ' the great ocean of woe resulting from misery under
the guise (apadesa = misa, LD. or chala, JV.) of the king's
prison.' The slight awkwardness observable when Sudraka
attempts to form long compounds, as in this verse (twice) and
in v. 24 (see note on that verse, above), serves at least to help
our appreciation of Bhavabhuti's exquisite skill in handling
them.

vi. I1. The use of visasane is curious; either we must under-
stand the word as an adjective (vinasajanake, JV. ; morderisch,
Bohtlingk), or else take the word as a locative of purpose (Vart-
tika on Panini, ii. 3. 36, quoted by LD. and the Calcutta com-
mentary) . The difficulty of taking the word as an adjective has
led me to adopt the second interpretation, as does Regnaud also ;
but the case does not fall exactly under the Varttika, because
there is no karmasamyoga, i. e. the thing sought after is not

436 A. W. Ryder, [1906.

connected with the object of the action (see Kielhorn's Gram-
mar, §633 a).

vi. 2. The last two lines are puzzling; the difficulty lies
partly in the sequence of thought, partly in the word gamya
'approachable.' For 'approachable' may mean either 'capable
of being pacified ' (gatva sandheyah santvaniya ity arthah, JV . ;
sarvesam sevyah, P. and the Calcutta commentary) or ' capable
of being attacked ' : so Bohtlingk (einem Konige kann man wohl
beikommen), and my translation. Besides, gamyo may repre-
sent agamyah out of sandhi. Then who is the balavant — Palaka,
or Fate, or Aryaka ? And finally, does daivi siddhih mean
'success (in attaining the throne) due to fate,' or is it merely
a circumlocution for daivam ? The translation of Bohtlingk>
which mine closely resembles, gives a reasonable sense, but
involves an awkward shifting of the point of view. Regnaud
is ingenious, but hardly convincing; he takes daivi siddhih to
mean ' fate,' and regards fate as identical with the king and the
powerful one ; the change of gender makes this very harsh. I
would suggest another interpretation, without very much con-
fidence in its correctness. "Even success (in attaining the
throne; rajapraptir api, JV.) cannot be avoided; a king (i. e. I
myself, destined to become king) must be appeased; for who
can fight with him who is powerful ? " That is, he cannot pre-
vent my becoming king, for fate wills it; he had better make
terms, for I am potentially more powerful than he. This inter-
pretation is rather subtle and tortuous for Sildraka ; but it saves
api from being a mere verse-filler, and preserves the same logi-
cal subject throughout the verse. On the other hand, it seems
irrelevant to the following prose.

vi. 3. The reading adattadando, found in Parab and JV.,
gives better meter than the anayatargalam of Stenzler and
Godabole.

vi. 6. I think we have to take visattha (visvastah) as a voca-
tive: 'my trusty men.'

vi. 7. The word sahasu (sabhasu) probably refers here to
gambling dens, dives.

vi. 151. It seems probable to me that purvavairl and purva-
bandhuh mean ' enemy in a former life ' and ' friend in a former
life' rather than ' former enemy ' and 'former friend.' In the
play itself, there is no indication of any previous acquaintance
of Aryaka's with Viraka or Candanaka.

Vol. xxvii.] Notes on the Mrcchakatika. 437

vi. 16. In Parab's text, read eka- for eka-.

vi. 181. There is no other allusion in the play to the fact
that Sarvilaka had ' given life ' to Candanaka.

vi. 202. See note on iv. 29.

vi. 22. I think it can hardly be doubted that Parab's chaya
interprets correctly the puzzling word kuccaganthi- as = San-
skrit kurcagranthi-. The word then means 'beard-knots,
scrubby beards.'

vi. 23s. In Parab's text, tatha should be printed as part of
the stage-direction. See Shankar Pandit's edition of the Mala-
vikagnimitra, page 168, note 51.

Act vii.

vii. I3. In Parab's text, read vaddhamanao for vaddhamanaa.

vii. 2. Parab's reading karmantojjhita- seems preferable
both to Stenzler's karmantotthita- and the vartmantojjhita- of
-Godabole and JV. ; for the -utthita- of Stenzler must be rather
forced to give a meaning, and so must the -anta- (= madhye
LD. ; madhyabhage JV.) of the other texts. Parab's reading
means of course 'left at the end of work.'

vii. 3. For the fourth line, compare D. D. Cunningham,
Indian Friends and Acquaintances (New York, 1904), pp. 64—
65: "The order of events is this: when everything is ready and
a desirable nest has been chosen, the cock-koil, conspicuous in
his shining black plumage and crimson eyes, seats himself on a
prominent perch, whilst the hen, in modest speckled grey garb,
lurks hidden among dense masses of neighbouring foliage. He
then lifts up his voice and shouts aloud, his voice becoming
more and more insistent with every repetition of his call, and
very soon attracting the attention of the owners of the nest,
who rush out to the attack and chase him away. Now comes
the chance for his wife, who forthwith nips in to deposit her
«gg. Very often she does this successfully before the crows
have returned, but every now and then she is caught in the act
and driven off like her husband, uttering volleys of shrill out-
cries."

vii. 4. Parab's reading asmad vyasanarnavotthitam has
rather more authority than Stenzler's asmad vyasanan navotthi-
tam, and is a more forcible expression ; Godabole and J V. have
the same reading as Parab, though LD. seems to explain the

438 A. W. Ryder, [1906.

other reading. If Parab's text represents the original, we have
an example of loose grammatical structure, inasmuch as asmat
must modify the prior member of the compound : cf . above, p. 420.

vii. 5. JV. and the Calcutta commentary are quite right in
pointing out that the reading nigadayugmam contradicts the
ekacaranalagnanigadah of vi. O52 (shortly before vi. 1) and the
padagrasthitanigadaikapasakarsi of vi. 1 (also the caranan niga-
dam apanaya of vii. 62 and the nigadam of vii. 8 ; on the other
hand, the plural is used at vii. 63>4>s) ; no doubt the reading
nigadam ekam of Parab and JV. or the nigadapasam of the
Calcutta commentary is in itself preferable : but the reported
manuscripts all read nigadayugmam.

vii. 6&. The expression samgacchehi niadaim is very curious.
If this represents samgacchasva nigadani, as it seems to, both
the construction and the sense are difficult ; for nigadani ought
to be in the instrumental, and there seems no reason why Mai-
treya should say "be united with the fetters." Commentators
and translators are alike unsatisfactory. It is just possible that
there is a smutty pun in samgacchasva, that Maitreya means
to hint that Carudatta, not being able to be united (sexually)
with Vasantasena, must be content to be united with what has
actually come in the cart, namely the fetters. But this does
not explain the case of niadaim.

vii. 61". The word before gatih is given in the following
forms; atilaghusamcara, alaghusanicara, alaghusamvara (Goda-
bole reads laghusamcara). Doubtless alaghusanicara gatih yields
a good meaning most easily : ' ' your progress is one whose move-
ment is not easy," i. e. 'you will find walking difficult'; but the
lectio difficilior alaghusamvara (Parab's reading) has a good
deal of authority. If it is correct, it seems to mean 'whose
concealment is* not easy'; in this case, Carudatta means that
Aryaka would probably be detected if he left the cart. The
reading -samcara may have crept in from the -samcare in the
next clause.

vii. 7. It is a not uncommon stylistic device in Indian
dramas to divide a verse. Sometimes the different parts are
spoken by different characters, as here and at Uttararamacarita
i. 33; Malatimadhava iii. 18; x. 8: sometimes the same charac-
ter speaks the whole verse, but is interrupted by prose speeches
from others, as at i. 44; Mudraraksasa vi. 16; Ratnavali iv. 19;

Vol. xxvii.] Notes on the Mrcchakatika. 439

Venisaihhara vi. 16 ; Prasannaraghava v. 35. A peculiarly
elaborate case is TJttararamacarita iv. 24—25, where a verse is
interrupted by a prose speech and by a second verse, this latter
verse being itself divided between two characters.

Act viii.

viii. I1. There is some doubt about the reading, but that
given in Parab's chfiya (. . . kevalam . . . saranam asrai) has
the most authority and is easier than . . . narah . . . sarane
(gacched iti sesah LD.). With the accepted reading, the line
means: "Having seen (mundane things: samsaram JV.) from
the standpoint of transitoriness, I am now the abode of virtues
only."

viii. 2. In the third line, we should probably read a (ca) for
Parab's ka (kva) ; then the jena of the first line governs lines
1—3. If we read ka (kva), it must mean, I suppose, 'in whom.'

viii. 33. Apte (s.v. apa-vah) gives to apavahayati in this
passage the meaning 'cause to carry the yoke,' while JV. gives
it the meaning 'beat' (tadayati). At any rate, it seems as if
the causative should mean a little more than 'chase away'
(Bohtlingk), especially in consideration of the case of gonam.
For the accusative shows that the action of the verb as well as
that of the gerund should be appropriate not only to the monk,
but also to the bullock. At viii. 441 we have a parallel passage,
in which vfihitah (P., however, takes vahide to represent badhi-
tah) is used without apa; here too, JV. explains vahitah by
taditah.

viii. 34. The word apanaka is used in the same sense of
' drinking party ' at Nagananda iii. 23. For the red radish,
compare the note on i. 52, above.

viii. 4. Bohtlingk's interpretation of the fourth line is, I
think, correct: " (der Garten . . .) kann wie ein neu angetre-
tenes Konigthum genossen werderi, ohne dass man es sich erst
zu erobern brauchte " ; but his interpretation of upabhogya (in
the PW.) as a noun seems unnecessary. It is more natural to
take anirjitopabhogyam as a karmadharaya ' to be enjoyed with-
out having to be conquered (by one's own efforts).' *.B^rhap<>
LD. is right in thinking that the vita intends to rebtke Sarii-
sthanaka's brutal conduct by contrasting it with the genDle invi-
tation of the park.

440 A. W. Ryder, 11906.

viii. 4". The word dhanya means 1. blessed, 2. infidel;
punya means 1. virtuous, 2. a brick watering-trough. This
accounts for Samsthanaka's blunder in thinking that he has
been called a materialist (carvaka) and a brick trough (kosthaka) ;
but why he should add kumbhakara, I am unable to see. The
word is omitted by some authorities and is not present in LD.'s
comment.

viii. 412. In Parab's text, delete the marks of punctuation
after pianti and nhaami ; for tahim must refer back to jahim.
The reading -savalaiih (-sabalani) of Stenzler and JV. seems
preferable to the -savannaim (-savarnani) of Parab and Goda-
bole, and has considerable authority. The expression in the
last clause ' I will make you a man of one blow ' seems very
idiomatic; in x. 35" we have the expression ekkappahalena
malia.

viii. 5. The phrase duram nigudhantaram is very puzzling
and the commentators are unsatisfactory. Perhaps it modi-
fies vastrantam and means ' (the hem of the garment) by which
the middle part is quite (duram) concealed', that is, he has
thrown the end so clumsily over his shoulder that it hides the
greater part of the garment: but this seems very awkward.
Assuming this explanation, the last two lines may be literally
translated : ' ' and he has not learned the (proper) arrangement
of the yellow robe ; and the hem of the garment, by which the
middle part is quite concealed, loose because of the bagging of
the cloth, does not fit on his shoulder."

viii. 6. We should expect vrksamansaih, to correspond in
formation with silasakalavarsmabhih ; the epithets are curious
enough.

viii. 10. This is one of the rare cases in which Samstha-
naka's mythology is correct.

viii. 14". Instead of ayam ftgatah, we should expect idam
agatam, since pravahana is neuter in 'Sanskrit; perhaps ayam
refers to Sthavaraka.

viii. 14". The expression hagge attanakelake na huvissarh is
unusual; it seems to mean 'I shall not be my own any longer,'
'I shall be dead.'

viii. 17'. The quotation is the last line of i. 31.

viii. 20. JV. takes dasanahuppalamandalehim as a bahuvrihi,
'whose lotus-heaps are ten finger-nails' (dasa uakha utpalaman-

Vol. xxvii.] Notes on the Mrcchakatika. 441

dalany utpalasamuha yayos tabhyam) ; but it seems more natu-
ral to take it as a karmadharaya, ' having ten finger-nails and
discs like those of lotuses.' In the next line, cadusadatadana-
is a karmadharaya; JV. analyzes catusatani priyavacanasatani
'va tadanani praharah.

viii. 22. JV. gives to the words tusti kadum the meaning
'to do me a favor,' which is ingenious, and possibly correct.

viii. 224. Here gandha means 'a mere smell,' 'a particle';
cf . the kasika on Panini v. 4. 136 : alpaparyayo gandhasabdah.
The same use of the word is found in Yogabhasya i. 48: na
tatra viparyasagandho 'py asti and iv. 15 : na 'nayoh sankara-
gandho 'py asti; while Regnaud and Bohtlingk see the same
meaning in raktagandhanuliptam in x. 3 : compare the note on
that verse, below. Regnaud has called attention to the mean-
ing 'a certain perfume' (canda) which the PW., on the author-
ity of Amara and the Medini, gives for raksasi, and sees a
deliberate pun on the part of the author; the suggestion is both
ingenious and convincing. JV. adds a further point by sug-
gesting that Samsthanaka misunderstands the vita's use of
akaryam ; the vita means ' something that must not be done,' ' a
sin,' but Samsthanaka takes him to mean 'something that can-
not be done,' an impossibility, and so declares that it is not a
witch after all.

viii. 24. Compare Manu viii. 86.

viii. 24'. Read palihissam for pahilissam.

viii. 28'. The word mallakka- (if, indeed, this be the correct
spelling) here and at ix. 5" has caused a good deal of trouble.
The ' earlier commentary ' (pracinatika) quoted by P. gives it
the meaning ' a small vessel made of a leaf ' (patraputika), and
this is adopted by LD. JV. reads gallakka- and offers the
meaning ' cur ' (kukkura) , but he quotes no authority, and on
ix. 521 he says that gallarka is a dialectic word for wine-vessel.
In Maitreya's speech at the very beginning of act i, and in
v. 6a we have the same word, with the, same Ms. variations
between initial m and initial g, and between single and double
k; in both places it must mean some kind of dish, and in v. 6s
it must mean 'a drinking- vessel '; and that is doubtless the
meaning which we have to accept in this passage.

viii. 29. This verse is repeated at ix. 7, with nitaram for
sutaram.

442 A. W. Ryder, [1906.

viii. 30*. ' The Prakrit nasena represents Sanskrit nyasena
and also nasena. The pun is obvious.

viii. 31. The words sevaam and kastamaa are doubtful.
Stenzler prints se vaam as two words, but in his chaya gives te
vayam ; te would of course be de in Prakrit, and this Bohtlingk
conjectures (p. 204). But the authorities speak overwhelm-
ingly for sevaam, Sanskrit sevakam. Then Stenzler's chaya
understands kastamaa as equal to Sanskrit kasthamayah rather
than kastamayah. Of course the t speaks against this, but the
meaning to be extracted from the words (te vayam kasthama-
yah) would then have to be that given by Bohtlingk, "Are we
to you men of wood ? " This seems an unnatural rendering in
itself, and has no support in the context. Certainly Vasanta-
sena understands the speaker to make a comparison between
himself and Carudatta, to the disadvantage of the latter. Her
sevitavyah (vs. 33) takes up the sevakam of the present verse,
and her daridrah (vs. 33) refers back to kastamayah. Samstha-
naka asks her why she does not desire him, and why she prefers
a poor man ; she answers that the poor man's character is good,
while his is bad. Probably, then, Ave must read kim sevakam
kastamayJi manusyah, and render: "why are poor men the
object of (your) devotion ? " True, the matter would be simpler
if we could read sevyante or sevyah.

viii. 32. Stenzler prints the entire first line as one word;
JV. divides khala caritanikrsta jatadosah, though his comment
offers the option between this division and that found in Parab's
text. Godabole, as also P. and LD., prefer the division found
in Parab. That this is the intention of the author is made
probable by the parallelism between khalacarita and sucarita-
caritam.

viii. 332. Of course the palasa and the kimsuka are the same ;
the blunder on Samsthanaka's part is like that found in the last
line of i. 41. I do not believe that the author intends a pun on
the name of the dempn Palasa, as P. and LD. say; Bohtlingk
(p. 204) argues effectively against this view.

viii. 34. I have taken some liberties with this verse in my
translation, in an effort to preserve something of the grim
humor of this critical scene. I am afraid that it is rather risky
to assume that the author, in using the name Dhundhumara,
plays- on the other meaning of the word, namely the insect
called indragopa.

Vol. xxvii.] Notes on the Mrcchakatika. 443

viii. 35 b. Parab's chaya should read mriyasva garbhadasi
mriyasva. In Sanskrit, as in English, we lose the assonance of
mala gabbhadasi mala following sumala gabbhadasi sumala in
viii. 35".

viii. 36. This verse and the next contain numerous difficul-
ties. The second line means: "who (really) came to her death
(kala-) when, being in love, she came (thinking) to sport with
him when he had come." In order to justify the accusatives
of the first two lines, we must supply hatvfi, as the Calcutta
commentary and JV. observe. In the fourth line, the chaya in
Parab and Godabole renders nisase by nihsvasa, and it is this
rendering which is represented by my translation ; but in Stenz-
ler and JV. the chaya has nihsvase, and that is of course what
we should expect from the Prakrit form of the word. The
phrase then means : " (Why do I boast of my strength of arm ?)
She dies merely at my breathing." The chaya in Parab, Goda-
bole, and JV. takes amba to represent Sanskrit amba, nomina-
tive, and this is precisely what we should expect; amba or
ambika 'mother' is used as a term of endearment, so e. g. at
viii. 1715. But the short final vowel of the Prakrit makes a diffi-
culty, and this difficulty is not avoided by the reading of Stenz-
ler's chaya, ambasmara.

viii. 37. The third line is desperate, so desperate that Reg-
naud does not attempt to translate it. Probably madeva repre-
sents mate 'va (not matai 'va : Stenzler) , as Parab's chaya has
it; the iva probably goes with draupadl (draupadisadrsi mata,
JV). We may tentatively translate the line: "my brother was
disappointed of his honor, and my father, and my mother (who
in this respect is) like that Draupadl." This translation assumes
the word-division sevavancida bhaduke, as Parab prints; but
the line seems nearly hopeless.

viii. 37". This speech of the vita's is very strange indeed,
and I do not see that the matter is helped by the reading pada-
yoh for padapah. The speech illustrates Bohtlingk's excellent
observation (Vorwort, p. i) : "Als eine Eigenthtimlichkeit ist
. . . auch dieses hervorzuheben, dass er . . . den Zuhorer
oder Leser . . . auf bevorstehende wichtige Begebenheiten vor-
ber<eitet und dadurch die Ueberraschung zwar einigermaassen
abschwacht, auf der anderen Seite aber auch die Neugier in
hohem Grade reizt." But it seems as if this end were attained
in the present case with unwonted awkwardness.

444 A. W. Ryder, [1906.

viii. 3718. For the expression, compare i. 30".

viii. 38. JV. takes daksinyodakavahini as an adjective
modifying ratih and suggests that the ' own region ' (svadesa) is
the south (daksina), because it is well known that rivers run
south. On the feminine form asraye, see above, p. 420.

viii. 40. This matter is printed by Stenzler as prose, by the
other editions as a verse ; if it makes a verse, as seems most
probable, the readings of the other three editions are nearer the
intent of the author than those adopted by Stenzler, since these
latter destroy the meter. The text is desperately bad. In the
second line, the editions all read savodiam (or sabo-), but the
explanations differ widely. Parab's chaya has savodinam and
P. says that a vodi is a coin of less value than a karsapana ; Stenz-
ler's chaya reads pustim and JV.'s saposanam; Godabole's chaya
reads savestikam and LD. explains vestika as meaning either
' turban ' or ' loin-cloth ' ; in this explanation he agrees with the
Calcutta commentary. One is tempted to prefer to all these
readings and interpretations the reading of Stenzler's Ms. B. :
sakodiam (sakotikam) : in this case, Samsthanaka is made to say:
"I'll give you wealth a hundred-fold, (I'll give you) a gold-
piece, I'll give you a penny, (I'll give you) ten millions." In the
third and fourth lines, my translation f olio ws^ Parab's chaya,
except that it is necessary to read samanyakam to agree with
dosasthanam: "Let this heroism of mine be a cause of censure
common to (all) men," a roundabout way of saying "Let the
perpetrator of the deed remain unknown." But there is rather
more authority for the reading of Godabole (with which JV.
practically agrees) : dusaddana phalakkame = duhsabdanam
phalaki-amah. Then the two lines mean: "Let this continued
reward of evil words (due) to me be common to (all) men."
The two readings thus give, at bottom, about the same sense.

viii. 42. I have taken jano 'yam in the ordinary sense of
ayam janah: "I think myself unworthy, etc." JV. takes it to
mean 'the average man' (sadharanamanava), and the transla-
tors take it similarly ; very likely they are right.

viii. 43. -sampanne: voc. fern., JV. ; loc. neut., Regnaud,
Bohtlingk.

viii. 437. I understand annam as a Sanskrit anyam, modify-
ing velam understood. Stenzler's chaya has anyas (supply alam-
karah), the other editions have ajna; yet Parab's punctuation
seems to indicate that he understands the Prakrit as I do.

Vol. xxvii.] Notes on the Mrcchakatika. 445

viii. 43V\ As Bohtlingk points out (p. 205), we probably
should read -kavodavaliae (as in i. 511) instead of -padolikae.
JV. attempts to explain the reading of the Mss., but his
attempt serves to confirm the suspicion that the Mss. are wrong :
prasadasya brhadattalikaya balayam abhinavanirmitayam agra-
pratolikayam pradhanarathyayam.

viii. 46. I have taken pattra in the meaning 'leaf in each
of its three occurrences in the verse ; this seems to be the under-
standing of P., who says pattrany eva, 'like the leaves they are.'
But the Calcutta commentary, LD., and JV. give to the words
vistirnapattrani . . . pattrani 'va the meaning 'like birds
whose wings are spread out ' ; it would be hard to find another
instance of pattra meaning 'bird.' Bohtlingk takes a middle
course in his translation: "diese ausgebreiteten Blatter regen
sich, so meine ich, wie Federn hin und her." It is perhaps
impossible to decide which interpretation is correct; the only
thing that is certain is that there is a play on the word pattra.

viii. 466. According to LD. and JV., the fact that Vasanta-
sena remembers the monk but does not remember her own bene-
faction to him, shows the nobility of her nature.

viii. 47. This matter (hattha- . . . niccale) is printed by
Parab as a verse ; also by JV. , who however gives it no verse-
number. The nature of the matter (cf. note on iv. 29, above),
and its position at the end of the act, make it a priori probable
that it does form a verse. The text printed by Parab scans 15.
17 : 12. 18 ; the last two lines form half of a regular arya. If we
read, with Stenzler and Godabole, hatthasanjadamuhasanjada-,
we obtain the scheme 13. 17: 12. 18. Thus we have the correct
number of syllabic instants, which are irregularly distributed in
the first half of the verse. In spite of this irregularity, it
seems most probable that we have to do with a stanza in the
arya meter.

Act ix.

ix. 1. In the fourth line, the reading of Parab and Godabole
does not scan correctly; if the first word is to be read gandhav-
vehi, it seems as if the second should be suvihidehim. This is
the text reproduced in my translation ; but I have taken gandh-
avvehi as the representative of Sanskrit gandharvaih, ' with
gandharvic, well-turned limbs." The fact that the Gandharvas

446 A. W. Ityder, [1906.

are male creatures and the persons mentioned in the third line
female, need not trouble us, as the blunder may be attributed
to Samsthanaka's ignorance. The reading of Parab and Goda-
bole is better supported than the gandhavve via suhidehirii of
Stenzler (with which JV. practically agrees). The latter reading
also gives a good sense, if we may take suhitaih to mean subhu-
sitaih (JV.) or sobhitaih (Calcutta commentary), or as the repre-
sentative of sukhitaih (Bohtlingk, page 205).

ix. 2. In Parab's text, khala- is apparently a misprint for
khana-. I think it is better to take mukke (muktah) in the
sense of 'hanging loose' (bandhanad bhransitah, JV.) than in
the sense of 'pearls' (Regnaud, Bohtlingk).

ix. 21. In kivinacestiam (bis) there is perhaps a pun; the
word means of course ' a wretched business,' but also perhaps
'a worm's business,' with reference to the kidaena above. The
possibility that krpana may here mean ' worm ' is increased by
the reading kimina- (apparently = krmina-) given by four Mss.
and by P.

ix. 3. In commenting on the third line, JV. says that the
king's judgment is confused by the exaggerations of the two
parties, that he is therefore likely to decide a case wrongly, and
that then he is subject to the penalty set forth in Manu viii.
128 (disgrace and hell).

ix. 4. In the second line, I have translated as if nasta dhru-
vam were the beginning of a new principal clause, but I am not
at all certain that this is correct.

ix. 5. In the last line, dvarbhave is puzzling. JV. takes it
as a locative absolute, supplying sati, ' there being an expedient ' ;
Bohtlingk interprets similarly. I have taken it as two words :
'a door (dvar, nom.) to truth,' but this is very doubtful.
G-odabole's Ms. K. has the reading dvabhyam vai, which is much
easier : ' (his heart devoted to others' interests) in behalf of both
parties (plaintiff and defendant)'.

ix. 521. For mallakkappamanaha, cf. note on viii. 281.

ix. 7. This verse is repeated from viii. 29, with nitaram for
sutaram. Parab, Godabole, and JV. print sphita for sphitah;
this perhaps indicates that the reading vipine (given by a major-
ity of the Mss.) for suksetre was the original reading, and that
suksetre has crept in from viii. 29.

ix. 77. J V. explains the curious word paasapindalakena thus :
payasapindam dugdhapakvam annam paramannam ity arthah,

Vol. xxvii.] Notes on the Mrcchakatika. 447

tad rcchati prapnotl 'ti tena payasannalobhine 'ty arthah : paya-
sannapraptaye lobhad yatha kriyate tatha maya 'pi 'ty arthah.

ix. 79. With the Calcutta commentary, I take -sthana in
moghasthanaya as an abbreviation for alamkarasthana ; compare
sunnaiih aharanatthanaim in ii. 20*.

ix. 11. I take ghoram asamsayam as a little clause by itself:
' the dreadful thing is certain.'

ix. 14. P. seems to have read cintamarga-. I have followed
LD. in taking duta- to mean 'attorney.' In the third line
-vasaka- is doubtless used with a double meaning. In reference
to the herons, it means 'screaming,' and in reference to court-
officers, it means 'slanderers, pettifoggers.' LD. has vasakah,
sabdarii kurvanah karnejapah pisuna eva; similarly JV., who
says: vasakah sabdam kurvana dhanaprataranartham vacanaca-
turah khala eva. I have adopted the reading -ruciram, which
seems better than Parab's -racitam.

ix. 19. The verse is desperately hard, and no comment or
translation is satisfactory. My translation aims to make sense,
but does violence to the text. JV. makes the sense-connection
between lines 2 and 3 by saying : casagrapakso hy upari varivar-
sanena malinibhavati tava mukham tu tadabhave 'pi malinam
drsyata iti bhavah. Accepting this, we may translate the verse
thus: "You are not, like the wing-tip of the casa, thoroughly
wet by the waters of the clouds in the sky; (yet it seems so,
because) this (accusation is) false — for (see!) this face of yours
attains lacklusterness like the winter lotus." But this is sadly
unsatisfactory.

ix. 22. The same conceit of leaving the ocean bare of gems
by reason of great riches occurs in the Meghaduta, in the verses
following i. 31 (regarded by Mallinatha as spurious).

ix. 23. The analysis which P. gives of the long compound
is to be preferred to that of the other interpreters : padapraha-
rena paribhava akramah sa eva vimanana taya baddhaguruka-
vairasya.

ix. 24s. The present participle viluppantam (vilupyamanam)
does not seem to correspond to the facts of the case; the read-
ing viluppam (viluptam) given by some Mss., seems preferable.

ix. 24&. Probably there is a little pun in lokavyavaharasya,
which may mean 'the conduct of men,' or 'a law-suit in the
world.' This I have tried to indicate in my translation.

448 A. W. Jtyder, [1906.

ix. 29. The word paravyasanena causes difficulty. LD. (fol-
lowed by Regnaud) interprets ' (although beset) by terrible mis-
fortune ' : parena vyasaneno 'palaksito 'pi ; similarly the Calcutta
commentary. J V. interprets ' with mere childish amusements ' :
parena kevalena vyasanena balyasulabhena kridanena. Boht-
lingk adopts this unusual meaning for vyasana, and accepts the
alternative reading bata for para, which is mentioned by the
Calcutta commentary and JV. I have taken paravyasanena to
mean 'with the misfortune of another,' but this is certainly
doubtful.

ix. 29". The translators have, I think, missed the point of
imassa. Of course, this masculine form cannot refer to Vasan-
tasena, in spite of the chaya in Parab and JV. The little clause
means: "it was right (for her) to give him the jewels (to stop
his crying, LD) but not (for me) to receive them."

ix. 30. Compare ix. 38.

ix. 304. There should be a mark of punctuation after hetu-
bhutah.

ix. 307. I have taken aniso (which is not found in all the
Mss.) to mean 'not master (of himself), mad'; but JV. explains
it as aksamo daridra ity arthah. The Prakrit bhandaa may
represent Sanskrit bhanda (chaya in Stenzler, Godabole, and
JV.) or bhanda (chaya in Parab, and P.); if the former be
intended, then kidajanadosabhandaa must be a compound, mean-
ing ' receptacle of crimes imputed to people ' ; if the latter, we
may take the expression as two words (or as a karmadharaya ;
so P.) meaning 'imputer of crimes to people, and buffoon.'
The latter seems preferable to me.

ix. 33. The last pada is found also in Kumarasambhava ii.
32. Whether this fact is or is not of importance in determin-
ing the relative dates of Kalidasa and Sudraka, I do not venture
to say.

ix. 35 '. I have taken the first two words as an impatient
exclamation : I do not believe that we have a play on words, as
LD. and Regnaud suggest. My view is perhaps supported by
JV., who prints abharanani a-, without sandhi.

ix. 36. JV. and Bohtlingk takes the fourth pada to mean
that the wishes of the speaker will fall to the ground (be disap-
pointed) when the lashes fall on Carudatta ; Wilson and I have
understood the pada to mean that the lashes -descend together

Vol. xxvii.] Notes on the Mrcchakatika. 449

with (in accordance with) the wishes of the speaker. I think
now that the former interpretation is the better; a similar play
on the root pat is found in ix. 31.

ix. 38. Compare ix. 30. In the third pada, Parab's reading
is excellent ; but we must take stri ratnam as two words. Then
the pada means: "a woman, and especially a jewel (of a
woman)."

ix. 391. As Bohtlingk points out (p. 209), the tti ought to
stand at the end of the speech.

ix. 411. Although LD. says that ambam refers to Carudatta's
mother, it seems more probable that it refers to his wife, Roha-
sena's mother; for there is no reference elsewhere in the play to
the mother of Carudatta.

Act x.

x. i. The difficulties of this verse are diminished if we can
regard kalana as the representative of the Sanskrit karanam
'pain.' We may then translate: "What then! Do not con-
sider (kalaya == vicaraya, JV.) the pain; being adepts in the
new-fangled managing of executions and fetterings, we are skil-
ful in cutting off heads and impaling in short order."

x. 3, JV. explains raktagandha- by raktacandana-, and
Regnaud's note (iv. 87) has the same suggestion. Bohtlingk
takes -gandha- in the sense of 'trifle,' as above at viii. 224, 241S.
I have supposed the word raktagandhanuliptam to contain a
rather mixed, but striking, metaphor, ' anointed with the odor
of blood.' Of these three interpretations, that of JV. and
Regnaud is perhaps the best.

x. II2. This speech is quoted at Dasarupa i. 46 (ed. Parab)
and at Sahityadarpana 384; in both places there are many, but
unimportant variants.

x. 12. Quoted at Dasarupa i. 46; ii. 4; Sahityadarpana 384,
with two variants; line 1, yat for me; line 3, nidhana- for
marana-. The commentary on the Sahityadarpana passage
explains nibidacaityabrahmaghosaih as follows: nibidani lokair
akirnani yani caityani: caityam ayatanam tulye ity Amarah:
pujadyayatanasthanani tesu ye brahmaghosa vedavadas taih.

x. 121. The reading udvijya can hardly be a mere blunder,
as it is explained by P. , but it is surely inferior to the udviksya
of the other texts.

VOL. xxvii. 30

450 A. W. Ryder, [1906.

x. 14. JV. reads asuvannamandanaaih as one word, arid
explains it as an adverb 'without any golden ornament'; but
this reading hurts the meter.

x. 17. Much better than the -bhojanam of all the editions is
-bhajanam, conjectured by Bohtlingk (p. 209), and read by two
of Godabole's Mss.

x. 19. Stenzler and JV. are doubtless right in reading
padicchidam (pratistam). JV. glosses the word with yathabhi-
lasitam.

x. 20. The first line is desperate. My translation is based
on Parab's;text, accepting P.'s explanation of pradesah by angani,
and presupposes the following literal translation: "his limbs are
parched (i. e. he is as good as dead) ; why need he conduct him-
self with bended head? " This is obviously most unsatisfactory.
Somewhat better are the readings and suggestions of JV., who
agrees in part with LD. and the Calcutta commentary. He
reads : sukkha vavadesa se kim panamia matthae na kaavvam
and explains : asya carudattasya .... vyapadesah kulanama-
dayah suska luptah? asya ca kim gunadikam ity arthah pra-
namya mastake na kartavyam? api tu sarvam eva pranatya
sirodharyam ity arthah: atha va vyapadesa vasantasenavadhaja-
nitapavadah suska mithyatvad aropita ity arthah.

x. 25. I interpret the second line as follows: " in which this
death is actually (api) a gain." In the fourth line, the reading
tvaya of Stenzler and Godabole is better than the maya of
Parab and JV. (but cf. the maya in x. 33).

x. 251; 282; 291. For pasadabalaggapadolika, cf. note on viii.
439' ", above.

x. 26. Cf. x. 38.

x. 27. For the construction, see above, p. 420.

x. 31. It is possible to separate niskaranopagata bandhava,
as Bohtlingk does, or, with JV., to take the word as a karma-
dharaya.

x. 328. I have followed Stenzler's chaya and Bohtlingk in
taking sankhalena as the representative of Sanskrit srnkhalena ;
but LD. and JV. translate it by sankhalena (san-), which they
explain to mean 'a drum-stick.'

x. 331. There is nothing to show whether Carudatta here
repeats ix. 30 or ix. 38.

Vol. xxvii.] Notes on the Mrcchakatika. 451

x. 338. In my translation, I have substituted the name Goha
for its synonym Viraka, as this is the only place where the
latter is used.

x. 35. I think that Bohtlingk is right (p. 211) in making
vesam = Sanskrit vesah. Cf. also JV., who says: vesa iva
paricchada iva.

x. 356. The matter from utthanta- to attanaam is printed
by Godabole and JV. as a verse. The nature of the material
makes it probable that they are right, though our text (11. 17:
10. 17) does not quite fit the scheme of an.arya.

x. 3510. In spite of the tradition, Bohtlingk is probably right
in thinking (p. 212) that eavasonnada represents ekaparsvonnata ;
but it does not seem necessary to substitute, with him, nauh
(as at ii. 202) for vasumdhara.

x. 3519. The phrase uttane bhavia must mean 'lie flat,' not
* stehe gerade ' (Bohtlingk) or ' tenez-vous bien raide ' (Reg-
naud). The word uttana means 'supine,' not 'erect;' besides,
the erect position would be unnatural for the operation pro-
posed. Then, at x. 403, Carudatta stands up (sahaso 'tthaya),
which he could not do if he were already erect; and Vasanta-
sena, who had fallen on his breast at x. 37', rises (utthaya) at
x. 401.

x. 38. Cf. x. 26.

x. 41. Here vidya means 'a spell for bringing the dead to
life' (LD., JV.),- i. e. vidya sanjivani, as it is called in the
Mahabharata.

x. 43. It is interesting to note that Jimutavahana, in the
fourth act of the Nagananda, uses the red marriage garments as
the insignia of death. Perhaps this passage and Mrcch. x. 43
stand in some connection with each other; if so, we have a sug-
gestion for the placing of the Mrcchakatika.

x. 46. Very likely there is a pun in the word sesabhutam;
the word may mean 'last,' and also 'being sacrificial flowers.'

x. 47. Stenzler's reading surareh is supported by only one
Ms., so that the reading balareh is doubtless correct. LD.,
JV., and the Calcutta commentary explain vasudhadhirajyam
as a bahuvrihi ; ' in which there is sovereignty over the whole
world.' The last half of the verse means then: "he has
obtained the entire kingdom of his enemy, implying sovereignty
over the world, like the kingship of Indra."

452 A. W. Ryder, [1906.

x. 471. Bohtlingk suggests (p. 213) that we read ayi for api
(apikaro 'tra prasne, JV.).

x. 48. The reading nirlkse is surely better than niriksye.

x. 481. The words atha va should be printed as part of the
text, as in the editions of Stenzler and Godabole.

x. 51". The authorities read without exception tatrabhavan;
but it seems as if we must change it to atrabhavan.

x. 53*. I take paurah, with the other editions, as part of the
stage-direction.

x. 54B-579. For the sake of completeness, I give a translation
of Nilakantha's interpolation, which may be inserted between
lines 23 and 24 on page 174 of my translation.

(Loud outcries are heard behind the scenes.)
Voices behind the scenes. See! The wife, the lady wife of
noble Carudatta thrusts back her little son, who clings at every
step to her garment's hem. The tearful bystanders would pre-
vent her, yet she mounts the blazing pyre.

Sarvilaka. (Listens and looks toicard the back of the stage.)
Ah, Candanaka! what does this mean, Candanaka? (Enter
Candanaka. )

Candanaka. Do you not see, sir? A great crowd has
gathered to the south of the Royal Palace. The wife, the lady
wife of noble Carudatta thrusts back her little son, who clings
at every step to her garment's hem. The tearful bystanders
would prevent her, yet she mounts the blazing pyre. I said to
her: "Madam, yoiT must not act too hastily. The noble Caru-
datta lives." But when the heart is full of sadness, who will
listen, who will believe ?

Carudatta. (In distress.) Oh, my beloved! what would
you do, while I yet live ? (He looks up and sighs. )
Although thy life upon the earth,
My virtuous wife, seem little worth,
Yet joy in heaven thou canst not find,
If thou dost leave thy lord behind. 55

(He swoons.)

Sarvilaka. What madness is this?

Yonder we needs must be so soon,

And here her husband lies in swoon ;

Alas! we must confess it plain,

That all our efforts are in vain. 56

Vol. xxvii.] Notes on the Mrcchakatika. 453

Vasantasena. Oh, sir, come to yourself. Go and bring her
back to life. Otherwise a calamity will be begotten of this
want of steadfastness.

Carudatta. ( Comes to himself and rises hastily. ) Oh, my
beloved, where are you ? Give me answer.

Candanaka. Follow me, sir. (All move about. Enter Caru-
datta's wife, as described; Rohasena, who clings to the hem of
her garment; Maitreya; and Radanika.}

Wife. (Tearfully.) Let go, my child. Do not hinder me.
I am fearful lest I hear of ill that happens to my lord. (She
rises, frees the hem of her garment, and moves toward the pyre.)

Rohasena. Oh, mother, think of me ! I cannot live without
you. (He runs up, and seizes again the hem of her garment.}

Maitreya. The sages declare it a sin for you, a Brahman's
wife, to mount the pyre without your husband's body,

Wife. Better to commit a sin than to hear of ill that happens
to my lord.

Sarvilaka. (Looks ahead.} She is near the flame. Hasten,
hasten! (Carudatta does so.}

Wife. Radanika, you must support my child, while I do
what I purpose.

Radanika. (Mournfully.) I too shall do what I have
learned from my mistress.

Wife. (Turning to Maitreya.} Then you must support
him, sir.

Maitreya. (Impetuously.} That your purpose may bear
fruit, a Brahman must take the lead in this action. And so I
shall precede you.

Wife. They both refuse me! (She embraces Rohasena.}
My child, you must care for yourself, that you may give us the
sesame and the water of sacrifice. Of what1 use are wishes,
when one is gone! (Sighing.} For my lord will not care for
you.

Carudatta. (Hears the words and hastens forward.} Yes, I
will care for my boy. (He raises Rohasena in his arms, and
clasps him to his breast.}

Wife. (Discovers him.} A miracle! I hear the voice of
my lord. (She looks more closely . Joyfully.} Thank heaven!
It is my lord himself. Now heaven be praised !

454 Ryder, Notes on the Mrcchakatika. [1906.

Rohasena. (Perceives his father. Joyfully.} Oh, oh! It
is my father that embraces me. (To his mother.} Mother,
now you are happy. Father will care for me. (He throws his
arms about Cdrudatta.}

Cdrudatta. ( To his wife. }

While he thou lovest more than breath
Was yet reprieved from jaws of death,

Whereto this mad emprize ?
Before the sun sinks in the west,
Why are the lotus' petals prest

Upon her sleeping eyes ? 57

Wife. My lord, it is just because she is so thoughtless that
she is kissed.

Maitreya. (Discovers Carudatta. Joyfully.} Hurrah!
These eyes see my friend. What power a faithful wife enjoys !
The mere purpose to enter the fire brings a reunion with her
love. ( To Carudatta. } Victory, victory to my friend !

Cdrudatta. Come, Maitreya! (He embraces him.}

Radanikd. What a wonderful providence! Sir, I salute
you. (She falls at Cdrudatta'sfeet.}

Cdrudatta. (Lays his hand upon her.} Rise, Radanika!
(He helps her to rise.}

Wife. (Perceives Vasantasend.} Thank heaven! My blessed
sister.

Vasantasend. Now am I blest indeed. (They embrace.}

Sarvilaka. Thank heaven! You live, with all your friends.

Cdrudatta. Yes, through your gracious aid.

x. 546. On pade, JV. says : pade pratipada ity arthah.

x. 567. On bhinnattanena, JY. has: bhinnatvena prthaktvena:
tad uktam Usanasa yatha : prthak citira samaruhya na vipra gan-
tum arhati : anyasam eva narinam strldharmo ' yam parah smr-
tah iti.

x. 5720. We should expect the dual: tau candalau . . .
bhavatam.

x. 59. Under stress of meter, I have omitted the words
kans cin nayaty akulan, which mean 'keeps some in suspense,'
and applies, like the other expressions of the first two lines,
both to people and to buckets.

The Buddhistic Rule Against Eating Meat. — By E. WASH-
BURN HOPKINS, Professor in Yale University, New Haven,
Conn.

THE fact asserted in Mr. H. Fielding Hall's People at School
(1906) that, although in the old days "it was immoral to take
life, wicked to eat meat and connive at butchery," it is now
the custom for Burmese Buddhists to do as they like in regard
to eating ("Everyone eats meat, even the monks," p. 257),
is explained by the author as a new departure, due to the stim-
ulating effect of the presence in Burma of the British beef-
eater. Is it not, in reality, a reversal in favor of a rule of
greater freedom ? Perhaps it is true that the Burman has but
lately found out for himself that the "religion of Necessity"
is better than the religion of Buddha as hitherto understood,
for a progressive Bui-man may have to eat well to compete with
British energy; but it is matter of interest to inquire just how
strict in ancient .times was the law against eating meat.

The great Protestant of India was no formalist. According
to the Vinaya, which seems rather to reflect the Master's atti-
tude than really to give his words as it pretends to do, Buddha
was perpetually harassed by imbecile friars, whose childish
questions he always answered in a spirit of liberality and com-
mon sense. Even later works show that to observe the spirit
and not the letter was the Buddhistic ideal. What is said of
verse may be applied to law,

attham hi natho saranam avoca
na bhyanjanam lokavidu mahesi,

" The all-wise Lord declared that salvation lies in the spirit and
not in the letter" (Comm. Khuddaka Patha, v). Thus, for
example, the general rule against suicide emanated from the
view that a saint ought to remain on earth as a good example ;
yet, in special circumstances, Buddha is represented as approv-
ing of suicide, as in the case of the Elder Godhika. Here it
is only the Evil One who objects to the act, on the ground that
to cut one's own throat argues a perfected saint (one indifferent

456 E. W. Hopkins, [1906.

to life), and that it is undesirable for the Evil One thus to lose
possession of the good Elder.1

So also the early Church, in the case of killing and eating,
appears to have been less strict than the later. The later Brah-
manic law, like that of the Jains, was very particular in regard
to these points. Except for sacrifice, to kill no sentient thing
and to eat no meat were absolute priestly laws. Even starva-
tion was barely an excuse for breaking these regulations, though
the class that did as it pleased despite the priests was reluctantly
conceded the right to hunt wild animals, and the priest even
found mythological reasons which made it meritorious for a
'king's man' to kill deer as well as men. People outside the
pale of respectability, fishers, fowlers, tanners, etc., were also
contemptuously permitted to remain in their odor of non-sanc-
tity. But for a priest even necessary agriculture was depre-
cated, ' because the plough hurts living things.' That this ' non-
injury ' rule was Buddhistic in origin is contrary to the evidence.
Even the oldest Brahmanic law, which is at least as venerable
as any Buddhistic literature, inculcates the general moral rule
of doing as one would be done by in the matter of injuring,
killing, and eating one's brother-animal.

Nevertheless, there are traces of a condition of things much
freer than this in the Brahmanic circle of a still earlier day. In
Ait. Br. iv. 3, man is said to eat, as well as rule over, cattle:
purusah pasusu pratisthito 'tti cai'nan adhi ca tisthati (on pasu
as implying cattle, cf. vi. 20). It is a Brahman priest who says
that he eats beef if it is off the shoulder (? ariisalam, Sat. Br. iii.
1. 2. 21). The common people are said at the same period to
be omophagous, amad (Kanva text, ib. iv. 5. 2. 16), and the
king has at least no scruples in regard to wearing leather san-
dals, varahya upanaha, ib. v. 4. 3. 19. Leather fastenings are
also alluded to in Ait. Br. v. 32. Brahman butchers are well
known, even in the Buddhistic period.2 The formal law-books
permit the eating of many animals, birds, and fishes, although
they denounce the sin of eating meat (see particularly Gautama,

1 For the rule, see the Patimokkha and Rhys-Davids on the Questions
of Milinda ; for the case of Godhika, Warren. HOS 3, p. 381.

2 Compare Jataka No. 495 (Fick, Soc. GHed, p. 141). Compare also
the casual allusion to a butcher-shop in Jat. No. 330. In Jat. No. 423 a
Brahman lives by hunting deer.

Vol. xxvii.] The Buddhistic Rule Against Eating Meat. 457

xvii, Baudh. xii, and Manu v). But perhaps casual allusions
reveal more than do the law-books. Convincing, for example,
is Tandya Br. xvii. 13. 9: "Clothed in a fresh garment he
comes up from the initiation-bath and during four months
neither eats meat nor has intercourse with a woman." That is
as much as to say, when not in a state of special purity one is
expected to eat meat. Compare Sat. Br. x. 1. 4. 13.

Similarly, although the Buddhist accepts and further promul-
gates, in his own decrepit dialect, the law " not to kill and not
to cause killing," it is evident that the law, if not late, was at
first not taken very strictly. Possibly, just as the Brahmanic
classes ('castes') were recognized, but without the Brahmanic
rigidity, which did not usually distinguish between letter and
spirit, so Brahmanic morality was, as an inheritance, not disre-
garded; but at the same time it was not so narrowly interpreted.
Among the many things which, according to the Buddhists'
scriptures, "people" (that is, non-Buddhistic people) objected
to in the conduct of the Buddhists was disregard of the life of
sentient beings. According to the same indisputable testimony,
people once found a Buddhist friar killing — of all animals — a
calf, and several times they complained that "followers of the
Bxiddha " hurt and killed living things. Even as an artistic back-
ground to the introduction of stricter rules, these tales, preserved
in the Buddhists' own books, can scarcely be supposed to be made
of whole cloth. There was some reason for the tale and for
the introduction of the more stringent rule. And the reason
was probably that, while Buddha really endorsed the rule Na
hanaye na ghataye, "Let one kill not, nor cause killing,"
neither he nor the early Buddhists interpreted it so strictly as
the Brahman was inclined to do. It is very seldom, for exam-
ple, that we find the addition "nor approve of others killing"
(Dhammika Sutta). To the Buddhist of the early days, meat
was not forbidden, though it was a work of supererogation to
abstain from it. Meat was a delicacy and it was not proper for
an abstemious friar to indulge, in any delicacies. On the other
hand, to take a vow not to eat meat was unusual ; it was dis-
tinctly an extra effort in ' acquiring merit.' ' The house-holder is

1 The Patimokkha prohibits meat and fish merely on the ground that
they are delicacies. The rules for novices contain no injunction against
eating meat. On the early usage among the friars, see Professor Rhys-
Davids' Buddhism, p. 164.

458 E. W. Hopkins, [1906.

distinguished from the ascetic in this, that the latter has no wife
and does not destroy life, while the former has a wife and does
destroy life (Muni Sutta). The rule of the ' King of Glory ' is
not a narrow one against meat; it is one of extreme liberality,
'Eat as you have been accustomed to eat.'1 There is a whole
sermon devoted to the expansion of the text, * defilement comes
not from eating meat but from sin ' (Amagandha Sutta), which,
as it seems to me, rather implies that meat was pretty generally
eaten (though the practice was looked upon by the stricter sort
as culpable) than that it was not eaten at all. Buddha him-
self (perhaps) died of eating pork, the flesh of a wild boar, an
idea so abhorrent to later Buddhism that the words sukara-
maddava, 'boar-tender' (-loin ?) was interpreted either as a
sauce or as a vegetable eaten by a boar; some said bamboo-
sprouts, other said a kind of mushroom, although no sauce or
vegetable is known by the name of 'boar-tender.' 2

It is in the light of such facts as these that the oft-repeated
rule "not to keep a store of raw meat" is to be interpreted.
The rule is generally given in connection with other purely
sumptuary regulations, such as not to keep a store of raw rice,
and far from seeming to prohibit meat it appears to imply its
use, the real prohibition being not against meat (any more than
against rice), but against the possession of a superfluous store.
Thus in the Gandhara Jataka, No. 406, it is said that a store of
salt and sugar even for one day, punadiva, used to be con-
demned, but now Buddhists hoard even for the third day.

Notable examples of freedom in respect of eating meat are to
be found in the Mahavagga, which gives other illustrations of
liberality. Thus, as to the other, we are told that, in the
northern country, for Buddhists to bathe more than once a fort-
night is a sin, but in the southern country they may bathe more
frequently, because it is the custom of the country. Here there
is no climatic necessity for the change, since what is called

1 Literally, " Ye shall eat as has been eaten" (Mahasudassana Sutta).

- Compare the Questions of Milinda, iv. 3. 22 and the discussion as to
bamboo, mushrooms, or sauce, Sacred Books of the East, xxxv, p. 244.
Boar flesh is common village-meat. Compare what the pigs say in Jat.
No. 388 : mamsatthaya hi posiyamase, " we are fattened for our flesh "
(p. 289), and further references below, p. 462. Still, some plant-names
begin with ' boar-,' and Buddha ought to have the benefit of the doubt.

Vol. xxvii.] The Buddhistic Rule Against Eating Meat. 459

' northern ' and ' southern ' is practically in the same clime. A
still better case is afforded by the similar regulation as to cov-
erlets. In the northern and middle part of the country, because
it is there customary to have coverlets made of vegetable mat-
ter, the Buddhists are to follow this custom; but when they go
south, where (as in Ujjain) people use animal skins as coverlets,
there they may use animal skins — a tacit condonation of the
slaughter of animals. As a medicinal remedy the Buddhist may
take intoxicating liquors1 and the flesh and blood and fat of
bears, alligators, swine, and asses. But a rule found in the
same work, vi. 31. 14, goes much further than this and really
gives the gist of the whole matter in permitting the use of meat,
if not killed for the express purpose of feeding the Buddhist.
The same rule holds as to fish. The Buddhists may eat it if
they "do not see, do not hear, do not suspect" that the fish
was caught especially for their use (ibid.). Elephants' flesh
and that of horses may not be eaten in time of famine, but this
is because they are parts of the "attributes of royalty";2 nor
that of dogs and snakes, but because such meat is disgusting.
Absolutely forbidden at such a time is only the flesh of human
beings3 and of other carnivora (ib. vi. 23. 9 ff.).

In regard to hurting sentient things, Brahmanism holds theo-
retically that even trees, plants, and grasses are kinds of ani-
mals. They differ only in being stable (fixed) instead of mobile ;
but a long argument which I have cited elsewhere from the Great
Epic shows that plants really see, hear, feel, and smell, as well
as possess the more obvious sense of touch, and that, therefore,
they are living, conscious things, endowed like other animals

1 A century after Buddha's death the Buddhist church (according to
tradition, Cullavagga. xii. 1) discussed the question whether it was per-
missible to drink unfermented toddy. The Buddhist was a teetotaler,
as was (ordinarily) the Brahman priest, but in this regard the church as
a whole appears to have been much stricter than the orthodox Hindus
(not of the priestly caste), who have always been addicted to intoxi-
cants. Even Brahman priests, north of the Nerbudda, were rum-
drinkers. Baudh. I. 2. 4.

2 Compare Jataka No. 397, p. 322, assa nama rajabhoga, " horses are
kings' property."

3 Cannibalism has left its trace in India in the stories of flesh-eating
Yakkas and Pisacas, natives of the Gilgit region (Dr. Grierson, in JRAS.
Jan. 1906 ; Jataka, 537;.

460 E. W. Hopkins, [1906.

with their own part of the anima mundi. This, sociologically,
is the older view as contrasted with that of the Buddhists, who
hold that a tree, for example, is 'conscious' only as containing
a living being (a dryad). Plants in themselves possess only one
organ of sense (feeling). So there is naturally less horror of
injury to plant-life (as plant) among Buddhists than among non-
Buddhists (the Brahmans and their followers),1 though rebirth
as a plant is more a theoretical possibility than an actual proba-
bility to both parties of believers in Karma. According to a
rather late compendium of heresies, the Brahmajala Sutta, the
Buddhist recluses, despite the tightening bonds of conventional
friarhood, still continued to injure growing plants, though it
was wrong to do so, as it was wrong " to accept raw meat " and
to kill living things. This reveals that raw meat was accepted
often enough to make it worth while to animadvert upon the
practice. But even this Sutta (like the rules for novices) does
not prohibit the eating of meat.

In the Edicts of Asoka there are several injunctions against
cruelty, but it is ordered merely that (even for sacrifice) no ani-
mals be killed "in future," with a recommendation to respect
the sacredness of life. Yet it is evident from the Fifth Pillar
Edict that the killing of animals was not unusual. Certain ani-
mals in the twenty-seventh year of Asoka's reign were made
exempt from slaughter, as were "all quadrupeds which are not
eaten or otherwise utilized by man," a clear intimation that
previously the slaughter of animals was not uncommon and that
" the more complete abstention from injury to animate creatures
and from slaughter of living beings " was, as proclaimed in the
Seventh Pillar Edict, brought about by Asoka, that is, a couple
of centuries after Buddha's death.3

1 There is, unfortunately, no common name for the Brahmanized
horde as there is for the followers of Buddha. I have sometimes for the
horde used ' orthodox,' as the Brahmans (i. e. the priests) use heterodox
(' unbelievers') especially of the Buddhists ; but the orthodox were any-
thing but a united fold, though they called themselves all, as against
Buddhists, ' believers.' On plants as ' having only one organ,' see Maha-
vagga iii. 1. 2.

2 The Edicts, however, are not for Buddhists alone but for all the
realm and in this particular may be aimed against Brahmanic (now
heterodox ! see the last note) rather than Buddhistic practices. Never-
theless, as no party distinction is made it may be presumed that the
Buddhists also needed a stricter rule. In connection with Brahmanic
practices, it must be noticed that beef -eating in the Mahabharata, though
common, is confined to ceremonial (sacrificial) consumption.

Vol. xxvii.J The Buddhistic Itule Against Eating Meat. 461

The Jatakas contain numerous instances revealing great free-
dom in respect of flesh-eating. For example, the Bpdhisat as
Sakka, in the Kumbha-Jataka, forbids the use of intoxicants,
but permits the enjoyment of flesh (mamsodanam sappipayasam
bhufija; No. 512, p. 20). So in Jat. No. 528, p. 235, the Bo-
dhisat as a mendicant, mahabodhiparibbajako, eat the flesh of a
monkey, makkatamamsam khaditva, and uses its skin as a robe,
though only in order to inculcate a lesson. In its Sanskritized
form, in the Jatakamala, this monkey appears as an illusion
(perhaps because of the audience; much as the "fatted calf" is
discreetly omitted from another parable in India at the present
day) and the Bodhisat merely "removes a skin made by him-
self " and then wears it, after causing the flesh to disappear
(carma'paniya sesam antardhapayam asa; sa tannirmittam vana-
racarma bibhrat, etc. HOS. 1, p. 147, 1. 19). That the deer is
a warrior's natural food is admitted in a casual remark addressed
to a priest, Jat. No. 483, p. 273, annam migo brahmana khatti-
yassa; but though a king hunts it is meritorious to renounce
the sport and devote oneself to charity. In No. 504, p. 437,
the king hunts not only deer but wild boar, migasukaradayo
vadhitva, and eats broiled venison, angarapakkam migamam-
sam. In No. 315, the Bodhisat gets a wagon-load of venison as
a gift ; but he takes the hunter from his cruel occupation, lud-
daka-kamma. In No. 12, a king is persuaded to stop killing
deer and all other animals. To eat the flesh of a golden pea-
cock, moro, which gives eternal youth and immortality (ib. 159
and 491) is perhaps too great a temptation to allow of its being
cited as an example ; yet the peacock was not forbidden food
either to the Brahman (Baudh. I. 12. 7) or to the pre-Asokan
Buddhist (v. note, loc. cit. S.B.E). Jat. Nos. 451 and 496 reveal
that meat-eating is almost a matter of course, even on the part
of the Bodhisat, who in No. 199 eats beef, gomamsam; while
the forest-ascetic (No. 496, p. 371, st. 28Q) says "I eat meat,"
just as he speaks of eating jujubes, lotus, etc. : sakam bhisam
madhuih mamsam badaramalakani ca, tani abhatva bhunjami
atthi me so pariggaho. In the introduction to the Sulasa Jataka,
No. 419, we have a scene depicting a pleasure-garden, where
thieves and servants indulge in fish, flesh, and intoxicants, mac-
chamamsasuradmi, which shows the vulgar popularity of flesh-
food. But in No. 436 a noble lady of Benares is fed on ghee,

462 E. W. Hopkins, [1906.

rice, fish, and flesh (p. 527, 1. 22) by the demon who would
woo her. Compare No. 434, where meat is eaten as a dainty.
Large bags of leather,1 mahante cammapasibbake, to hold money,
are referred to in No. 336. Leather is used to make chariot-
harness (No. 22) and the clothing of a mendicant, cammasatako
paribbajako, in No. 324. Roast pig is used to celebrate a mar-
riage-feast (Nos. 30 and 286) and roast lizard is recognized as
good food (in No. 333) ; though it is a false Buddhist ascetic,
dussllatapaso, who in Nos. 138 and 325 is fond of such diet.
But crow's meat is sent (as earnest of better) to the Bodhisat by
the king in No. 214, and in No. 220 the scholiast tells a story
(to illustrate a Jataka verse) which implies that a king regularly
ate meat (animals might be slaughtered in Benares any day
except on fast-days). No. 241, p. 245, even notes the occasion
on which, according to tradition, men who had eaten all the
fresh meat they could, first began to dry it: tasmim kila kale
vallurakaranam udapadlti vadanti. A very good example of
the casual, matter-of-course way in which meat-eating is referred
to will be found in Jataka No. 106 (p. 417), wherein a young
man is advised by his father, the Bodhisat, not to marry, simply
because he will have to run errands for his wife: "When she
wants to eat fish or meat or has need of ghee or salt or rice,
etc." (and sends you to, do her errands), yada macchamamsadmi
va khaditukama bhavissati sappilonatanduladlhi va pan' assa
attho bhavissati. Here the worldly fat girl is imagined as eat-
ing meat as naturally as salt, etc.

The whole matter of meat-eating is epitomized in the verse,
ascribed to the Bodhisat in the Telovado Jataka (No. 246) :

bhunjamano pi sappaimo na papena upalippati,
that is, according to the context, if one who has divine wisdom
eats fish or meat, even when he knows it is prepared for him, he
does no wrong.2 Not meat-eating per se, not the fact that meat

1 The common use of leather, as Prof. Bloomfield remarked when this
paper was read, has been recently exemplified by excavations made in
the Northern deserts. Leather nooses are made in Jat. 206 (p. 153).

2 In the exaggerated language of the Bodhisat, one may even eat the
flesh of the donor's wife or child. Only the slayer is sinful, not the
eater. The comment is : samamsakam bhattam adasi . . . samano
Gotamo janam uddissa-ka^am mamsarh bhunjati, " He gave meat-food
. . . Gotama the ascetic knowingly eats meat prepared especially for
him." Buddha here accepts in full the precepts of the Bodhisat.

Vol. xxvii.] The Buddhistic Rule Against Eating Meat. 463

was prepared especially for the eater, not even the fact that the
latter knows of the circumstances, makes the eater guilty of sin.
But he must eat with no evil in the heart, no indulgence of
appetite.1 With the same liberality, which distinguishes the
ethics of Buddha from that of his ascetic rivals, we find the rule
that no evil Karma attaches to an act of unintentional wrong-
doing, as laid down in the Kuru-dhamma of Jataka No. 276 (p.
377), acetauakam kammam na hoti, the Brahmanic rule being that
there must be expiation for unintentional as for intentional sin.2
Devadatta, Buddha's rival, permitted no eating of flesh-meat;
Buddha permitted it with restrictions as to the spirit in which
it was eaten. In other words, early Buddhism was opposed
to this form of asceticism as to other austerities, which in
themselves are valueless.3

The great distinction between killing and eating may seem
rather pharisaical, but it existed. To kill an animal, to be
butcher, fowler, or fisher, was wrong, and to connive at slaugh-
ter in order to gratify appetite was also wrong.4 But when the
beast had been killed without prior connivance on the part of
the Buddhist the flesh might be accepted and eaten. The early
Buddhist seems to have thought that, as the animal was dead
anyway, he might as well make use of it and did not trouble his
conscience with questions of ' tainted ' offerings. If uncom-
monly ascetic he might refuse it as being a delicacy, but not
because meat as meat constituted sinful diet. Probably the
later accession of Brahmanical converts tended to the greater
strictness of the Buddhist in this regard, until he came to say

1 Compare the passage (cited by Mr. Rouse at this place in his transla-
tion) from Hardy's Manual, p. 327: "Those who take life are in fault,
but not the persons who eat the flesh. My priests [in contrast with
those of Devadatta J have permission to eat whatever is customary to
eat in any place or country, so that it be done without the indulgence
of appetite, or evil desire." The Cullavagga on this point, vii. 3. 14,
mentions only fish, but the contention is the same.

2 Compare with this No. 528 (p. 237) : akamakaramyasinim kuv-idha
papena lippati. For the Brahmanic rule, see JRAS. July, 1906, p. 584.

3 See the Majjhima Nikaya, pp. 77-8, for a catalogue of useless
austerities.

4 Compare Jat. No. 506 (p. 458), where the king-snake refuses to eat
frogs especially killed for him, with the idea "n'esa mam nissaya
maressatiti " (na khadati), " not for my sake shall he kill."

464 Hopkins, The Buddhistic Rule Against Eating Meat. [1906.

with St. Paul ' If eating meat my brother do offend I will eat
no more meat.' The theory of transmigration had, I imagine,
little to do with the matter either with Buddhists or with
Brahmans ; though Buddha admits that a man may be reborn as
an animal, for, in speaking of the death of a perfected saint, he
couples together, as the fruit of such saintliness, the destruction
of "hell and rebirth as an animal." The Jatakas, too, recog-
nize man's rebirth as a beast, but these are not of the earliest
Buddhistic era, and, generally speaking, the primitive Buddhist
is reborn as man and, if not, he is more likely to reappear as an
unfathered divinity in consequence of virtue than as an animal
in consequence of evil.1 At any rate, man's rebirth as an animal
(with a possible cannibalism) is never suggested as a reason why
a Buddhist should not eat meat, although the Brahmanic view
was that the animal later would eventually take revenge by
eating (in another life) the former eater. Yet even here the
idea is not that one should abstain from flesh through fear of
eating a reincarnated relative.

To take life, in distinction from eating meat, results in going
to hell or in rebirth either as an animal, a ghost, petti visaye, a
demon, asurakaye, or a human being of short life, appayukasam-
vattanikam (hotl ti, 'said the Bodhisat'), Jat. 55 (p. 275).

1 On the knotty question as to how a future Buddha could be born as
an animal, cf. Jatakamala xxxiii. st. 3. Despite his sufficient wisdom
dharmasanjni 'pi, he had acquired "bits of (evil) Karma," karmalesans
tans tan samasadya, which reduced him to a beast. The Bodhisat him-
self explains rebirth in animal form as due to neglect in a previous life to
perform good works (kusalakammassa akattata), as he says Jat. 31 (p. 205,
lines 1 and 7, to Sujata): tvam pana kusalam akatva tiracchanayoniyam
nibbatta. The same question arises in regard to the sins committed by
Bodhisattas, such as reverting to sensuality (Jat. 251), keeping and
knocking down his wife (No. 199), seducing a girl (No. 62), or even lead-
ing a band of robbers. In the last case the Jataka-maker ascribes such
faults rather vaguely " to the stars," nakkhattadosena, Jat. No. 279 (p.
389), apparently forgetful of the Bodhisat's own words, kim karissanti
taraka (No. 49, Nakkhatta-Jat.). Rather an interesting statement is made
in Jat. 431 (p. 499), to the effect that on some (unexplained) occasions,
ekaccesu th&nesu, Bodhisats may destroy life, commit adultery, and
drink intoxicants, sura ; but they may not tell deceitful lies, musavado,
which destroy the reality of things. Truth is the highest virtue. In
mediaeval Sk. literature abstention from meat is a sign of virtue, as in
the Hitopadesa, where, more specifically, eating meat " on the Lord's
day " (1. 3) is unlawful.

PEOCEEDINGS

AMERICAN ORIENTAL SOCIETY,

AT ITS

MEETING IN NEW HAVEN, CONN.
1906.

The annual meeting of the Society was held in New Haven,
Conn., on Tuesday and Wednesday of Easter week, April 17th
and 18th, in the Library of the Classical Club, in Phelps Hall.

The following members were present at one or more of the
sessions :

Allen, F. S.

Carus

Jackson, A. V. W.

Morris, Miss

Allen, Miss

Currier

Jackson, J. D.

Miiller

Arnold, W. R.

Gottheil

Jastrow

Oertel

Barton

Gray

Jewett

Palmer

Blake

Grieve, Miss L. C.

G. Lanman

Quackenbos

Blodgett

Haas

Lilley

Ropes

Bloomfleld

Harper

Lyon

Torrey

Boiling

Haskell

Moore, G. F.

Toy

Brown

Hock

Moore, J. H.

Van Name

Campbell

Hopkins

Moore, Mrs. G. F.

Ward, W. H.

• •

Total 40.

The first session began on Tuesday morning at eleven o'clock,
with President Gilman in the chair.

The minutes of the last annual meeting, held in Springfield,
Mass., April 27th and 28th, 1905, were read in abstract by the
Recording Secretary, having already been printed in the twenty-
sixth volume of the Society's Journal.

The Committee of Arrangements, through Professor Hop-
kins, presented its report in the form of a printed programme.
Professor Torrey announced that the Graduates' Club extended
its privileges to the members of the Society during their stay
in New Haven ; that a luncheon would be given by the New
Haven members of the Society on Wednesday at one o'clock,
VOL. xxvii. 31

466 American Oriental Society's Proceedings, April, 1906. [1906.

to the men, at the Graduates' Club, while the ladies in attend-
ance on the meeting would be entertained at the same time, at
the Country Club, by Mrs. Hopkins and Miss Whitney; and
that arrangements had been made for a dinner on Wednesday
evening at half past seven, at the Tontine Hotel.

The succeeding sessions of the Society were appointed for
Tuesday and Wednesday afternoons at half past two o'clock,
and for Wednesday morning at half past nine.

The Corresponding Secretary, Professor E. W. Hopkins,
reported as follows :

Letters of acceptance have been received from all those
elected to membership at the last Meeting. Greetings were
sent to Dr. James C. Hepburn and a reply received.

In response to an invitation from the Philosophical Society to
send a delegate to represent this Society at Philadelphia during
the Franklin Bicentenary exercises, President Gilman consented,
at the Secretary's request, to act as our representative.

In accordance with the order of the Directors, the Journal of
the Society has been copyrighted.

The Ethnological Survey P.' I. has been added to the list of
exchanges.

The death of the following members of the Society was
reported :

HONORARY MEMBERS.

Professor Jules Oppert.
Professor Friedrich von Spiegel.

CORPORATE MEMBERS.

Mr. Clarence H. Clark.

Dr. P. L. Armand de Potter.

Samuel Fales Dunlap.

President William Rainey Harper.

Rev. Lauren P. Wolfe.

MEMBER OF THE SECTION FOR THE HISTORICAL STUDY OF

RELIGIONS.

Professor George S. Goodspeed.

CORRESPONDING MEMBER.
Rev. Joseph Edkins.

Tributes were paid to Professor Oppert, by Professor Gott-
heil ; to President Harper, by Professor Toy ; and to Professor
Spiegel, by Professor Jackson.

The report of the Treasurer, Professor F. W. Williams, was
read by Mr. J. D. Jackson.

Vol. xxvii.J Report of Treasurer. 467

RECEIPTS AND DISBURSEMENTS BY THE TREASURER OF THE
AMERICAN ORIENTAL SOCIETY FOR THE YEAR

ENDING DECEMBER 81, 1905.

The Treasurer in presenting his report makes mention of the fact
that he has been obliged again this year to draw upon the reserves of
the Society by withdrawing $500 to meet the cost of the annual volume
from the accumulated interest of the Cotheal Publication Fund. The
Society in order to get upon a safe basis should either increase its income
or reduce its present expenditure by about three hundred dollars
annually.

RECEIPTS.

Balance from old account, Dec. 31, 1904, $ 772.53

Dues (179) for 1905 $895.10

" (47) for other years 235.18

" (19) for Hist. S. E. Sect 38.00

$1,168.28

Sales of publications 120.86

State Nat. Bank Dividends 108.88

Interest Suffolk Savings Bk 11.06

" Prov. Inst. for Savings 53.51

" National Savings Banks .34

Subscriptions collected for Subvention 95.00

1557.43

$2,329.96
EXPENDITURES.

T., M. & T. Co., printing vol. XXV11 .. $572.13

vol. XXVr 779.79

paper 87.78

" sundry printing 36.44

Binding 21.80

Library cards and stamp 13.75

Subvention to Orient. Bibliographic 95.48

Honoraria to editors 200.00

Librarian, postage and express 20.88

Treasurer, " 19.12

Balance to general account 482.79

$2,329.96
STATEMENT.

1904 1905

I. Bradley Type Fund(N. H. Savings Bank) $2,192.52 $2,297.44

II. Cotheal Publication Fund (Pr. Inst. Savings). 1,000.00 1,000.00

III. State National Bank Shares 1,950.00 1,950.00

IV. Life Membership Fund 225.00 300.00

V. Connecticut Savings Bank deposit. 5.52 5.52

VI. National Savings Bank deposit 10.50 10.50

VII. Accrued Interest in II 516.12 69.63

VIII. " " IV 69.98 81.04

IX. " " VI .34

X. Cash on hand.. -67.68 331.78

$6,037.32 $6,046.25

468 American Oriental Society's Proceedings, April, 1906. [1906.

The report of the Auditing Committee, Mr. J. D. Jackson
and Mr. A. P. Stokes, was presented by Mr. Jackson, and is as
follows :

April 16th, 1906.

In the absence of my colleague, Mr. Anson Phelps Stokes, Jr., I have
made the audit of the accounts of the American Oriental Society and
hereby certify that I have examined the account book of the Treasurer
of the society and have found the same correct, and that the foregoing
statement is in conformity therewith.

I have also compared the bills and vouchers and statement of balances
accompanying the same and have found them to be correct.

JOHN DAY JACKSON,

Auditor for the Society.

The Librarian, Professor Hanns Oertel, presented his report,
as follows:

REPORT OF THE LIBRARIAN.

In the past winter foundations were laid on the University Campus
for a new building planned as the beginning of an eventual complex of
University library buildings. It immediately adjoins the present Chit-
tenden Library and will occupy the space between the latter and the
Old University Library. The new building covers some fifty thousand
odd square feet, will accommodate at least four hundred thousand books
and will also contain the executive offices of the librarians and a num-
ber of consultation rooms. The main part of the building will consist
of a compact book stack of six tiers, arranged on the one hand to be
accessible to the various reading rooms and consultation rooms, and
also arranged that parts of the stack can be sectioned off and used for
the deposit of such works as from their value or character need not be
open to the general body of students for consultation. It is proposed
thus to create a section of the stack for the library of the American
Oriental Society, insuring the Society's books being kept together under
lock and key, and still readily accessible to its members. The stack
will of course be absolutely fire-proof, well lighted during the day by
natural light from the west, and by electric light during the evening .
hours. Well-lighted tables will be placed in the stack and in immediate
proximity to the books, where the members of the Society may consult
its books free from interruption. The expansion of the library will be
provided for, so that as long as the books of the Society are in deposit
in the University Library they will be kept together and separate from
the University's books.

The impending transfer of our books from their present cramped
quarters emphasized with particular force the need of cataloguing the
Society's Library ; for without a catalogue it is practically impossible
to administer properly or use conveniently a large collection of books
such as our library now holds. However, with the chronic deficit in
our treasury, such a plan would have remained a pious wish had it not
been for the most generous offer of Miss Margaret D. Whitney to under-
take the cataloguing of the Society's library without compensation and
as a labor of love. Miss Whitney, who is a trained librarian, undertook
the work with characteristic energy and I hope to be able to report at

Vol. xxvii.] Report of Librarian. 469

the next meeting that the greater part of the Society's library has been
competently catalogued. A new and heavy indebtedness will thus be
added to the many and important obligations which our Society, since
its inception, owes to the name of William Dwight Whitney. The
thanks of the Society are due to John C. Schwab, Esq., Librarian'of the
Yale University Library, for many favors and valued assistance, and to
Henry Gruener, Esq. , of the Yale University Library, who, as in pre-
vious years, has rendered much assistance to the Librarian.

Respectfully submitted.

HANNS OERTEL,
April 16, 1906. Librarian.

On motion of Professor Bloomfield the Society expressed its
thanks to Miss Margaret Whitney for her services in preparing a
catalogue of the library.

The report of the Editors of the Journal, Professors Hopkins
and Torrey, was presented by Professor Torrey, and is as fol-
lows:

REPORT OF THE EDITORS.

The twenty-sixth volume of the Journal was issued in two parts, the
First Half appearing in August, 1905. and the Second Half in March,
1906. The volume contained 468 pages in all ; or 416 pages exclusive of
the Proceedings, List of Additions to the Library (covering the period
April 1898-April 1905), List of Members, and Notices.

The late appearance of the Second Half of this volume was due
chiefly to a printers' strike, which continued for a long time and
reduced our publishers to straits. It is likely that the effect of this
same strike will be felt in the publication of vol. xxvii ; but probably
the delay will not be great.

The following persons, recommended by the Directors, were
elected members of the Society :

HONORARY MEMBERS.
Professor Ferdinand Justi. Professor Ignaz Goldziher.

CORPORATE MEMBERS.

Harold Herman Bender. Francis Branch Blodgett.

C. A. B. Brockwell. Wallace B. Fleming.

William L. De Vries. Edward Atwood Henry.

Florence Alden Gragg. Charles Ellsworth Home.

Paul Vincent Harper. Robert J. Lau.

Franklin W. Hooper. Alonzo A. Madsen.

Nicholas A. Koenig. Martin A. Meyer.

Isaac G. Matthews. Hans K. Moussa.

Rowland H. Mode. James B. Nies.

William Muss-Arnolt. Samuel G. Oliphant.

Charles J. Ogden. Johann F. Scheltema.

Benjamin W. Robinson. Gilbert Campbell Scoggin.

Mrs. Edward E. Salisbury. John M. P. Smith.

O. A. Toffteen. Eben F. Thompson.

470 American Oriental Society's Proceedings, April, 1906. [1906.

MEMBERS OF THE SECTION FOR THE HISTORICAL STUDY OF

RELIGIONS.

Rev. W. A. Shedd. Rev. G. E. White.

The committee appointed at Springfield to nominate officers
(Messrs. Moore, Jackson, and Jastrow: see Journal, vol. xxvi,
p. 425) reported through its chairman, Professor Moore, as
follows :

At the last meeting of the Society, Mr. Gilman, who has
filled the office of President most acceptably since 1893, having
expressed his desire that he should not be asked to serve after
this year, a committee was appointed to nominate officers at the
present meeting, with the understanding that the election would
be held at the first session, so that the President should take
his seat at the meeting at which he was elected; and it was
voted " that in future the President be requested to prepare an
address on some phase of the progress or significance of Oriental
studies to be read at the annual meeting." This Society has
been peculiarly fortunate in its Presidents, and it has been
accustomed to re-elect them from year to year so long as they
were willing to serve it. In most of the other American learned
societies the presidency is an honor which is annually conferred
upon some distinguished scholar, and it was plainly in the mind
of the Society in the plan' which it adopted at Springfield that
it should in future be so among us also. It is not proposed that
any new rule be made, but merely that the usage hitherto pre-
vailing shall not be regarded as having the force of prescrip-
tion.

The committee nominated the following officers, who were
unanimously elected :

President — Professor Crawford Howell Toy, of Cambridge, Mass.

Vice-Presidents — Dr. William Hayes Ward, of New York ; Professor
Charles R. Lanman, of Cambridge ; Professor Maurice Bloomfield, of
Baltimore.

Corresponding Secretary— Professor E. Washburn Hopkins, of New
Haven.

Recording Secretary — Professor George F. Moore, of Cambridge.

Secretary of the Section for Religions — Professor Morris Jastrow, Jr.,
of Philadelphia.

Treasurer — Professor Frederick Wells Williams, of New Haven.

Librarian— Professor Hanns Oertel, of New Haven.

Directors — The officers above named ; and President Daniel Coit Gil-
man. of Washington ; Professor Robert F. Harper, of Chicago ; Pro-
fessors Richard Gottheil and A. V. Williams Jackson, of New York ;
Professor Paul Haupt, of Baltimore ; Professor Henry Hyvernat. of
Washington ; Professor Charles C. Torrey, of New Haven.

Vol. xxvii.J Communications. 471

Professors G. F. Moore, A. V. W. Jackson, and M. Jastrow,
Jr. were appointed a committee to nominate officers at the next
annual meeting.

At twelve o'clock President Gilman delivered his address, on
"Recent Contributions by Americans to the Knowledge of the
Orient."

The second session of the Society began 'at half-past two
o'clock, with President C. H. Toy in the chair, and proceeded
to the reading of papers. The following communications were
presented :

Dr. F. R. Blake, of Johns Hopkins University, Contributions
to comparative Philippine grammar.

Dr. J. C. Ferguson, of Shanghai, China, The abolition of
competitive examinations in China.

Professor Gottheil, of Columbia University, Muhammad
Abdu, late Mufti of Egypt. — Remarks were made by Professor
W. M. Mtiller.

Dr. L. H. Gray, of Newark, N. J., The Dutangada of
Subhata, now first translated from the Sanskrit and Prakrit.

Mr. G. C. O. Haas, of Columbia University, Dhanika's com-
mentary on the Dasariipa. — Remarks were made by Professor
Lanman.

Professor Hopkins, of Yale University, On the Buddhist rule
against eating meat. — Remarks were made by Professors Bloom-
field, Lanman, and Toy.

Professor Boiling, of the Catholic University of America, A
plan for an edition of the Atharva Veda Parisistas. — Remarks
were made by Professor Bloomfield.

Professor Jackson, of Columbia University, Indo-Iranian
notes (read by title) ; The Zoroa'strians of Yezd.

Professor Jastrow, of the University of Pennsylvania, Did
the Babylonian temples have libraries? — Remarks were made by
Professors W. M. Mtiller, Barton, and Gottheil.

Professor Jewett, of the University of Chicago, Announce-
ment of a new series of Arabic texts.

At a quarter past five the Society adjourned to half past nine
Wednesday morning.

The Society met on Wednesday morning at half past nine
o'clock, with President Toy* in the chair. Professor Toy
expressed his appreciation of the honor the Society had done in
electing him to the Presidency;, in following a line of eminent
predecessors he should rely on the co-operation of the members
of the Society to maintain its high standard of efficiency.

The reading of papers was resumed. The following com-
munications were presented:

Professor Lanman, of Harvard University, Twin consonants
at the junction of two consecutive words. — Remarks were made
by Professors Jewett, Jastrow, and Boiling, and Dr. Ward.

472 American Oriental Society's Proceedings, April, 1906. [1906.

Professor Moore, of Harvard University, On a leather gar-
ment from an Egyptian tomb, now in the Boston Museum of
Fine Arts, which has been described as an " ephod." — Remarks
were made by Professor W. M. Milller and Dr. Ward.

Professor Bloomfield, of Johns Hopkins University, Some
emendations of Vedic texts. — Remarks were made by Professor
Lanman.

Dr. Lucia C. G. Grieve, of New York, Demonstration of a
Brahman woman's daily worship in the house.

Miss Margaretta Morris, of Philadelphia, Race and custom
in the Malay archipelago. — Remarks were made by Professor
Jastrow.

Mr. J. H. Moore, of Columbia University, Onomatopoetic
words in Sanskrit. — Remarks were made by Professor Bloom-
field.

Professor Fay, of the University of Texas, Studies of San-
skrit words (presented by Professor Bloomfield).

Professor W. M. Milller, of Philadelphia, The goddess of the
Carthaginians. — Remarks were made by Professors Gottheil,
Torrey, Jastrow, and Moore.

Professor Torrey, of Yale University, The last sentence of
the Tabnit tomb inscription. — Remarks were made by Profes-
sors Gottheil and Arnold.

Professor Haupt, of Johns Hopkins University, Semitic verbs
derived from particles; The etymology of Hebrew mohel, "cir-
cumciser;" The Talmudic passage, Succah 455, and Psalm
cxviii. 27; Some Indo-European etymologies (read in abstract
by Dr. Blake).

Dr; Lau, of Columbia University, 'abel in the Bible; A sup-
plement to the Old Babylonian vocabulary (read by title).

At half past twelve the Society took a recess till half past
two.

The Corresponding Secretary reported for the Directors that
the next annual meeting will be held in Philadelphia, Penn.,
beginning on April 4, 1907. A committee of arrangements was
appointed, consisting of Dr. Talcott Williams, Professor Jas-
trow, and Professor Mtiller.

The Directors further reported that they had appointed Pro-
fessor E. Washburn Hopkins and Professor Charles C. Torrey
Editors of the Journal for the ensuing year.

On motion of Professor Hopkins the following resolution was
unanimously adopted by a rising vote :

The members of the American Oriental Society desire to
express to their retiring President, Dr. Daniel C. Gilman, their
appreciation of the services rendered during the thirteen years
of his Presidency, and to record herewith their heartiest thanks
for the zeal and devotion he has always shown to the interests
of this Society.

Vol. xxvii.] Communications. 473

The following communications were presented:

Professor Jastrow, of the University of Pennsylvania, The
story of Tabi-Utul-Bel.

Professor Lyon, of Harvard University, Abbreviated legal
expressions in early Babylonian contracts; Seal impressions .on
early Babylonian contracts. — Remarks were made by Messrs.
Jastrow, Arnold, Gottheil, and Allen.

Professor Gottheil, of Columbia University, A distinguished
family of Fatimide Cadis (al-Nu'man) in the tenth century.

Professor Lyon, of Harvard University, The date of the
Babylonian king Immeru. — Remarks were made by Messrs.
Allen and Jastrow.

Professor Toy, of Harvard University, The decay of totem-
ism. — Remarks were made by Professor Hopkins.

Dr. F. R. Blake, of Johns Hopkins University, Comparative
syntax of the noun and its modifiers in Semitic ; The expression
of case by the verb in Tagalog (read by title and abstract).

The following resolution of thanks wTas unanimously adopted:

The American Oriental Society desires to express its thanks
to Yale University for the use of the Library of the Classical
Club as a meeting place ; to the Graduates' Club for the priv-
ileges of its Club house ; to the resident members of the Society
and the ladies for their hospitality ; and to the Committee for
the arrangements, which have contributed much to the success
of the meeting and the pleasure of those in attendance.

At five o'clock the Society adjourned, to meet in Philadelphia,
Penn., April 4th, 1907.

The following communications were read by title :
Mr. Aaron Ember, Modern additions to the Hebrew language ;
A new Modern Hebrew-English and English-Hebrew dictionary.
—Dr. T. C. Foote, The metrical form of the Songs of Degrees,
—Mr. L. B. Wolfenson, The Pi'lel in Hebrew; The infixes la.
li, and lo in Tagalog. — Professor Christopher Johnston, Egyp-
tian chronology. — Professor D. G. Lyon, Female votaries in
the days of Hammurabi. — Professor Oertel, Contributions from
the Jaimimya Brahmana, sixth series ; A Greek inclusive (ellip-
tical) plural and a Sanskrit inclusive singular.

474 American Oriental Society's Proceedings, April, 1906. [1906.

LIST OF MEMBERS.

EEVISED, DECEMBER, 1906.
The number placed after the address indicates the year of election.

I. HONORARY MEMBERS.

M. AUGUSTE EARTH, Membre de 1'Institut, Paris, France. (Rue Garan-

ciere, 10.) 1898.
Dr. RAHKRISHNA GOPAL BHANDARKAR, C.I.E., Dekkan Coll., Poona, India.

1887.

JAMES BURGESS, LL.D., 22 Seton Place, Edinburgh, Scotland. 1899.
Dr. ANTONIO MARIA. CERIANI, Ambrosian Library, Milan, Italy. 1890.
Prof. BERTHOLD DELBRUECK, University of Jena, Germany. 1878.
Prof. FRIEDRICH DELITZSCH, University of Berlin, Germany. 1893.
Prof. ADOLPH ERMAN, Steglitz, Friedricli Str. 10/11, Berlin, Germany.

1903.
Prof. RICHARD GARBE, University of Tubingen, Germany. (Biesinger Str.

14.) 1902.
Prof. KARL F. GELDNER, University of Berlin, Germany. (Luebecker Str.

40, N. W.) 1905.
Prof. M. J. DE GOEJE. University of Leyden, Netherlands. (Vliet 15.)

1898.
GEORGE A. GRIERSON, C.I.E., D.Litt., I.C.S. (retired), Rathfarnham,

Camberley, Surrey, England. Corporate Member, 1899; Hon., 1905.
Prof. IGNAZ GOLDZIHER, vii Holld-Utcza 4. Budapest, Hungary.
Prof. IGNAZIO GUIDI, University of Rome, Italy. (Via Botteghe Oscure,

24.) 1893.

Prof. FERDINAND JUSTI, Marburg, Germany.
Prof. HENDRIK KERN, University of Leyden, Netherlands. 1893.
Prof. FRANZ KIELHORN, University of Gottingen, Germany. (Hainholz-

weg, 21.) 1887.
Prof. ALFRED LUDWIG, University of Prague, Bohemia. (Konigliche

Weinbirge, Kramerius-gasse 40.) 1898.
Prof. GASTON MASPERO, College de France, Paris, France. (Avenue de

1'Observatoire, 24.) 1898.
Prof. THEODOR NOELDEKE, University of Strassburg, Germany. (Kalbs-

gasse 16.) 1878.
Prof. RICHARD PISCHEL, University of Berlin, Germany. (Halensee,

Joachim Friedrichstrasse 47.) 1902.
Prof. EDUARD SACHAU, University of Berlin, Germany. (Wormser Str.

12, W.) 1887.

Prof. ARCHIBALD H. SAYCE, University of Oxford, England. 1893.
Prof. EBERHARD SCHRADER, University of Berlin, Germany. (Kronprin-

zen-Ufer 20, N. W.) 1890.

Vol. xxvii.] List of Members. 475

Prof. JULIUS WELLHAUSEN, University of Giittingen, Germany. (Weber

Str. 18a.) 1902.
Prof. ERNST WINDISCH, University of Leipzig, Germany. ( Universitats

Str. 15.) 1890. [Total, 25.]

II. CORPORATE MEMBERS.

Names marked with f are those of life members.

Rev. Dr. JUSTIN EDWARDS ABBOTT, Tardeo, Bombay, India. 1900.

Dr. CYRUS ADLER, U. S. National Museum, Washington, D. C. 1884.

F. STURGES ALLEN, 246 Central St., Springfield, Mass. 1904.

Miss MAY ALICE ALLEN, 397 Elm St. , New Haven, Conn. 1906.

Prof. EDWARD V. ARNOLD, University College of North Wales, Bangor,

Great Britain, 1896.

Mrs. EMMA J. ARNOLD, 2?f> Washington St., Providence, R. I. 1894.
Prof. WILLIAM R. ARNOLD, Theological Seminary, Andover, Mass. 1893.
Dr. KANICHI ASAKAWA, Publishing Dept., Waseda University, Tokyo,

Japan. 1904.

Rev. EDWARD E. ATKINSON, City Hall, Taunton, Mass. 1894.
Hon. SIMEON E. BALDWIN, LL.D., 44 Wall St., New Haven, Conn. 1898.
LEROY CARR BARRET, Box 86, Johns Hopkins University, Baltimore, Md.

1903.

Prof. GEORGE A. BARTON, Biyn Mawr College, Bryn Mawr, Pa. 1888.
Prof. L. W. BATTEN, 232 East llth St., New York. 1894.
Prof. HARLAN P. BEACH, Yale University, 79 Howe St., New Haven,

Conn. 1898.
Prof. WILLIS J. BEECHER, D.D., Theological Seminary, Auburn, N. Y.

1900.

HAROLD H. BENDER, Johns Hopkins Univ., Baltimore, Md. 1906.
Rev. JOSEPH F. BERG, Port Richmond, S. I., N. Y. 1893.
Dr. WILLIAM STURGIS BIGELOW, 60 Beacon St., Boston, Mass. 1894.
Prof. JOHN BINNEY, Berkeley Divinity School, Middletown, Conn. 1887.
Dr. FRANK RINGGOLD BLAKE (Johns Hopkins Univ.), Dixon Park, Mt.

Washington, Md. 1900.
Rev. DAVID BLAUSTEIN, Educational Alliance, 197 East Broadway, New

York, N. Y. 1891.

FREDERICK J. BLISS, Clifton Springs^New York.

FRANCIS B. BLODGETT, General Theological Seminary, Chelsea Square,
t N. Y. 1906.
Prof. CARL AUGUST BLOMGREN, Augustana College and Theol. Seminary,

Rock Island, 111. 1900.
Prof. MAURICE BLOOMFIELD, Johns Hopkins University, Baltimore, Md.

1881.
Prof. CHARLES W. E. BODY (General Theological Seminary), 9 Chelsea

Square, New York, N. Y. 1897.
Dr. ALFRED BOISSIER, Le Rivage pres Chambe'sy, Switzerland. 1897.

476 American Oriental Society's Proceedings, April, 1906. [1906.

Dr. GEORGE M. BOLLING, Catholic Univ. of America, Washington, D. C.

1896.
Prof. JAMES HENRY BREASTED, University of Chicago, Chicago, 111.

1891.
Prof. CHAS. A. BRIGGS (Union Theological Seminary), 700 Park Ave.,

New York, N. Y. 1879.
Prof. C. A. B. BROCKWELL, Univ. of Kings College, Windsor, Nova Scotia,

Canada. 1906.

Dr. PAUL BRONNLE, 73 Burdett Ave., Westcliff-on-Sea, England. 1903.
Prof. FRANCIS BROWN (Union Theological Seminary), 700 Park Ave.,

New York, N. Y. 1881.

Prof. CARL DARLING BUCK, University of Chicago, Chicago, 111. 1892.
Rev. JOHN CAMPBELL, Kingsbridge, New York, N. Y. 1896.
Rev. SIMEON J. CARR, 1527 Church St., Frankford, Philadelphia, Pa.

1892.
Pres. FRANKLIN CARTER, care Hon. F. J. Kingsbury, Waterbury, Conn.

1873.

Dr. PAUL CARUS, La Salle, Illinois. 1897.

Dr. I. M. CASANOWICZ, U. S. National Museum, Washington, D. C. 1893.
Miss EVA CHANNING, Hemenway Chambers, Boston, Mass. 1883.
Dr. FRANK DYER CHESTER, United States Consulate, Buda-Pesth, Hungary.

1891.

Rev. HENRY N. COBB, 25 East 22d St., New York, N. Y. 1875. .
fGEORGE WETMORE COLLES, 62 Fort Greene Place, Brooklyn, N Y. 1882.
Prof. HERMANN COLLITZ, Bryn Mawr College, Bryn Mawr, Pa. 1887.
Miss ELIZABETH S. COLTON, Easthampton, Mass. 1896.
C. EVERETT CONANT, Translator-Interpreter Executive Bureau, Manila,

P. I. 1905.

WILLIAM MERRIAM CRANE, 16 East 37th St., New York, N. Y. 1902.
OSCAR T. CROSBY, Cosmos Club, Washington, D. C. 1904.
STEWART CULIN, Brooklyn Institute Museum, Eastern Parkway, Brooklyn,

N. Y. 1888.

Rev. CHARLES W. CURRIER, 941 F St., Washington, D. C. 1904.
Prof. JOHN D. DAVIS, Princeton Theological Seminary, Princeton, N. J.

1888.

LEE MALTBIE DEAN, Westbrook, Maine. 1897.

Prof. ALFRED L. P. DENNIS, 5735 Washington Ave., Chicago, 111. 1900.
JAMES T. DENNIS, University Club, Baltimore, Md. 1900.
Rev. WILLIAM L. DE VRIES, 301 A St. S. E., Washington, D. C. 1906.
Rev. D. STUART DODGE, 99 John St., New York, N. Y. 1867. «

DAVID J. DOHERTY, M.D., P. O. Box 727, Manila, P. I. 1905.
Dr. HARRY WESTBROOK DUNNING, 5 Kilsyth Road, Brookline, Mass. 1894.
WILBERFORCE EAMES, Lenox Library, 890 Fifth Ave., New York, N. Y.

1897.

Prof. FREDERICK C. EISELEN, Garrett Biblical Inst., Evanston, 111. 1901.
Mrs. WILLIAM M. ELLICOTT, 106 Ridgewood Road, Roland Park, Md.

1897.

Vol. xxvii.] List of Members. . 477

Prof. LEVI H. ELWELL, Amherst College, Amherst, Mass. 1883.

AARON EMBER, Johns Hopkins Univ., Baltimore, Md. 1902.

Rev. ARTHUR H. EWING, The Jumna Mission House, Allahabad, N. W. P.,

India. 1900.

Rev. Prof. C. P. FAGNANI, 772 Park Ave., New York, N. Y. 1901.
Prof. EDWIN WHITFIELD FAY (University of Texas), 200 West 24th St.,

Austin, Texas. 1888.

ERNEST F. FENOLLOSA, 159 Church St., Mobile, Ala. 1894.
Prof. HENRY FERGUSON, Trinity College, Hartford, Conn. 1876.
Dr. JOHN C. FERGUSON, 110 Bubbling Well Road, Shanghai, China.

1900.

RALPH HALL FERRIS, 41 East 69th St., New York, N. Y. 1905.
CLARENCE STANLAY FISIIER. Rutledge, Delaware Co., Pa. 1905.
•j-Lady CAROLINE DE FILIPPI FITZ GERALD, 167 Via Urbana, Rome, Italy.

1886.

Rev. WALLACE B. FLEMING, Maplewood, N. J. 1906.
Rev. THEODORE C. FOOTE, Johns Hopkins University, Baltimore, Md.

1900.

fFRANK B. FORBES, 65 Marlborough St., Boston, Mass. 1864.
Rev. JAS. EVERETT FRAME (Union Theological Sem.), 700 Park Ave., New

York, N. Y. 1892.
Prof. ISRAEL FRIEDLAENDER (Jewish Theological Seminary), 1549 St.

Nicholas Ave., New York, N. Y. 1904.

Dr. WILLIAM H. FURNESS, 3d, Wallingford, Delaware Co., Penn. 1897.
Dr. FLETCHER GARDNER, Fort Michie, N. Y. 1905.
ROBERT GARRETT, Continental Building, Baltimore, Md. 1903.
Prof. BASIL LANNEAU GILDERSLEEVE, Johns Hopkins Univ., Baltimore,

Md. 1858.

Pres. DANIEL COIT GILMAN, 614 Park Ave., Baltimore, Md. 1857.
Louis GINZBERG, 60 West 115th St., New York, N. Y. 1900.
Prof. WILLIAM WATSON GOODWIN (Harvard Univ.), 5 Follen St., Cam-
bridge, Mass. 1857.
Prof. RICHARD J. H. GOTTHEIL (Columbia Univ.), 2074 Fifth Ave., New

York, N. Y. 1886.

Miss FLORENCE A. GRAGG, 26 Maple Ave., Cambridge, Mass. 1906.
JACOB GRAPE, JR., Bond and Jefferson Sts., Baltimore, Md. 1888.
Louis H. GRAY, 354 Summer Ave., Newark, N. J. 1897.
Miss LUCIA C. GRAEME GRIEVE, 462 West 151st St., New York, N. Y.

1894.
Miss LOUISE H. R. GRIEVE, M.D., Satara, Bombay Presidency, India.

1898.

Dr. KARL JOSEF GRIMM, Pennsylvania College, Gettysburg, Pa. 1897.
Prof. Louis GROSSMANN (Hebrew Union College), 2212 Park Ave.,

Cincinnati, 0. 1890.

CHAS. F. GUNTHER, 212 State St., Chicago, 111. 1889.
Rev. ADOLPH GUTTMACHER, 1833 Linden Ave., Baltimore, Md. 1896.
GEORGE C. 0. HAAS, 64 East Seventh St., New York, N. Y. 1903.
Dr. CARL C. HANSEN, Lakawn Lampang, Laos, Siam (via Brindisi,

Moulmain, and Raheng). 1902.

478 American Oriental Society's Proceedings, April, 1906. [1906.

PAUL V. HARPER, 59th St. and Lexington Ave., Chicago, 111. 1906.

Prof. ROBERT FRANCIS HARPER, Univ. of Chicago, Chicago, 111. 1886.

Prof. SAMUEL HART, D.D., Berkeley Divinity School, Middletown, Conn.

Prof. PAUL HAUPT (Johns Hopkins Univ.), 2511 Madison Ave., Balti-
more. 1883.

Rev. EDWARD HAYES, Johns Hopkins Univ., Baltimore, Md. 1904.

Dr. HENRY HARRISON HAYNES, 6 Ellery St., Cambridge, Mass. 1892.

EDWARD A. HENRY, 70 Middle Divinity, University of Chicago, 111. 1906.

Col. THOS. WENTWORTH HIGGINSON, 25 Buckingham St., Cambridge, Mass.
1869.

Prof. HERMANN V. HILPRECHT (Univ. of Pennsylvania), 403 South 41st
St., Philadelphia, Pa. 1887.

Prof. FRIEDRICH HIRTH (Columbia Univ.), 501 West 113th St., New
York, N. Y. 1903.

Prof. CHARLES T. HOCK (Theological Seminary), 220 Liberty St., Bloom-
field, N. J. 1903.

fDr. A. F. RUDOLF HOERNLE, 8 Northmoor Road, Oxford, England. 1893.

Rev. HUGO W. HOFFMAN, 306 Rodney St., Brooklyn, N. Y. 1899.

Prof. FRANKLIN W. HOOPER, 502 Fulton St., Brooklyn, N. Y. 1906.

Prof. E. WASHBURN HOPKINS (Yale Univ.), 299 Lawrence St., New
Haven, Conn. 1881.

CHAS. E. HORNE, 5836 Drexel Ave., Chicago, 111. 1906.

Rev. THOMAS P. HUGHES, D.D., 585 Park Place, Brooklyn, N. Y. 1905.

Rev. ROBERT E. HUME, Ahmednagar, India. 1900.

Miss ANNIE K. HUMPHEREY, 1114 14th St., Washington, D. C. 1873.

HENRY MINOR HUXLEY, 31 William St., Worcester, Mass. 1902.

Prof. HENRY HYVERNAT (Catholic Univ. of America), 3405 Twelfth St.,
N. E. (Brookland), Washington, D. C. 1889.

Prof. A. V. WILLIAMS JACKSON (Columbia Univ.), 10 Highland Place,
Yonkers, N. Y. 1885.

JOHN DAY JACKSON, 86 Crown St., New Haven, Conn. 1905.

Prof. MORRIS JASTROW, JR. (Univ. of Pennsylvania), 248 South 23d St.,
Philadelphia, Pa. 1886.

Miss MARY JEFFERS, Bryn Mawr, Pa. 1900.

Rev. HENRY F. JENKS, P. O. Box 79, Canton Corner, Mass. 1874.

Prof. JAMES RICHARD JEWETT, Quadrangle Club, Chicago, 111. 1887.

Prof. CHRISTOPHER JOHNSTON (Johns Hopkins University), 21 West 20th
St., Baltimore, Md. 1889.

Prof. MAX KELLNER, Episcopal Theological School, Cambridge, Mass.
1886.

Miss ELIZA H. KENDRICK, 45 Hunnewell Ave., Newton, Mass. 1896.

Prof. CHARLES FOSTER KENT (Yale Univ.), 406 Humphrey St., New
Haven, Conn. 1890.

Prof. GEORGE L. KITTREDGE ( Harvard University ) , 9 Hilliard St., Cam-
bridge, Mass. 1899.

NICHOLAS A. KOENIG, 80 West 12th St., New York, N. Y. 1906.

Rev. GEORGE A. KOHUT, 781 West End Ave., New York, N. Y. 1894.

STEPHEN HERBERT LANGDON, Grassistrasse 26, Leipzig, Germany. 1902.

Vol. xxvii.J List of Members. 479

fProf. CHARLES ROCKWELL LANMAN (Harvard Univ.), 9 Farrar St.,

Cambridge, Mass. 1876.

ROBERT JULIUS LAU, 650 Leonard St., Brooklyn, N. Y. 1906.
BERTHOLD LAUFER, American Museum of National History, 77th St., and

Central Park West, New York, N. Y. 1900.
f HENRY C. LEA, 2000 Walnut St., Philadelphia, Pa, 1898.
C. S. LEAVENWORTH, care of Brown, Shipley & Co., 123 Pall Mall, London,

England. 1900.

FREDERICK LENT, 177 Lawrence St., New Haven, Conn. 1906.
Prof. CASPAR LEVIAS, Hebrew Union College, Cincinnati, Ohio. 1892.
ROBERT LILLET, Graf ton, Mass. 1894.

Prof. THOMAS B. LINDSAY, Boston Univ., Boston, Mass. 1883.
Prof. CHARLES E. LITTLE (Vanderbilt Univ.), 308 Gowday St., Nash-
ville, Tenn. 1901.

Prof. ENNO LITTMAN, University, Strassburg, i/E., Germany. 1902.
Rev. JACOB W. LOCH, 89 Schermerhorn St., Brooklyn, N. Y. 1899.
PERCIVAL LOWELL, care of Putnam &, Putnam, 50 State St., Boston, Mass.

1893.

fBENJAMiN SMITH LYMAN, 708 Locust St., Philadelphia, Pa. 1871.
Prof. DAVID GORDON LYON, Harvard Univ. Semitic Museum, Cambridge,

Mass. 1882.
ALBERT MORTON LYTHGOE, Metropolitan Museum of Art, New York,

N. Y. 1899.

Mrs. MATILDA R. McCoNNELL, 112 Madison Ave., New York, N. Y. 1890.
Prof. DUNCAN B. MACDONALD, Hartford Theological Seminary, Hartford,

Conn. 1893.
Rev. CHARLES S. MACFARLAND, 37 Bayview Ave., South Norwalk, Conn.

1898.
Lieut. WILLIAM E. W. MACKINLAY, 1st U. S. Cavalry, Lemon Building,

1729 New York Ave., N. W., Washington, D. C. 1904.
ALBERT A. MADSEN, Durham, Conn. 1906.
Prof. HERBERT W. MAGOUN, 70 Kirkland St., Cambridge, Mass.
Prof. MAX L. MARGOLIS, Hebrew Union College, Cincinnati, Ohio. 1890.
Prof. ALLAN MARQUAND, Princeton Univ., Princeton, N. J. 1888.
Prof. WINFRED ROBERT MARTIN, Trinity College, Hartford, Conn. 1889.
ISAAC G. MATTHEWS, McMaster Univ., Toronto, Canada. 1906.
MARTIN A. MEYER, 22 St. Francis Place, Brooklyn. N. Y. 1906.
TRUMAN MICHELSON, R.F.D. 48, Ridgefield, Conn. 1899.
Mrs. HELEN L. MILLION (nee LOVELL), Hardin College, Mexico, Missouri.

1892.
Prof. LAWRENCE H. MILLS (Oxford University), 119 Iffley Road, Oxford,

England. 1881.
Prof. EDWIN KNOX MITCHELL (Hartford Theol. Sem.), 57 Gillette St.,

Hartford, Conn. 1898.

ROLAND H. MODE, 144 South D, Univ. of Chicago, 111. 1906.
Prof. J. A. MONTGOMERY (P. E. Divinity School), 6806 Green St., German-
town, Pa. 1903.

480 American Oriental Society's Proceedings, April, 1906. [1906.

Prof. GEORGE F. MOORE (Harvard University), 3 Divinity Ave., Cam-
bridge, Mass. 1887.

JUSTIN HARTLEY MOORE, 8 West 119th St., New York, N. Y. 1904.

fMrs. MARY H. MOORE, 3 Divinity Ave., Cambridge, Mass. 1902.

PAUL ELMER MORE, 265 Springdale Ave., East Orange, N. J. 1893.

Miss MARGAUETTA MORRIS, 2106 Spruce St., Philadelphia, Pa. 1903.

Prof. EDWARD S. MORSE, Salem, Mass. 1894.

Rev. HANS K. MOUSSA, Burlington, Wis. 1906.

Rev. Dr. PHILIP S. MOXOM, 83 Dartmouth Terrace, Springfield, Mass.
1898.

Prof. W. MAX MUELLER, 27 North Farson St., Philadelphia, Pa. 1905.

Rev. Prof. A. J. ELDER MULLAN, S.J., Georgetown University, Washing-
ton, D. C. 1889.

Mrs. ETHEL WATTS MUMFORD, 31 West 81st St., New York, N. Y. 1904.

Dr. WILLIAM MUSS-ARNOLT, Belmont, Mass. 1906.

Rev. JAS. B. NIES, Christ Church Rectory, Sharon, Conn. 1906.

Prof. CHARLES ELIOT NORTON, Cambridge, Mass. 1857.

Rt. Rev. Mgr. DENNIS T. O'CONNELL, D.D. (Catholic University), Wash-
ington, D. C. 1903.

Prof. HANNS OERTEL (Yale Univ.), 2 Phelps Hall, New Haven, Conn.
1890.

CHAS. J. OGDEN, 250 West 88th St., New York, N. Y. 1906.

Miss ELLEN S. OGDEN, St. Agnes School, Albany, N. Y. 1898.

SAMUEL G. OLIPHANT, Johns Hopkins Univ., Baltimore, Md. 1906.

Prof. PAUL OLTRAMARE (University of Geneva), Ave. de Bosquets, Ser-
vette, Geneve, Switzerland. 1904.

fRoBERT M. OLYPHANT, 160 Madison Ave., New York, N. Y. 1861.

JOHN ORNE, 104 Ellery St., Cambridge, Mass. 1890.

Prof. GEORGE W. OSBORN, New York University, New York, N. Y. 1894.

Rev. Dr. CHARLES RAY PALMER, 562 Whitney Ave., New Haven, Conn.
1900.

Prof. LEWIS B. PATON, Hartford Theological Seminary, Hartford, Conn.
1894.

Prof. WALTER M. PATTON, Baker Univ., Baldwin, Kansas. 1903.

Dr. CHARLES PEABODY, 197 Brattle St., Cambridge, Mass. 1892.

Prof. ISMAR J. PERITZ, Syracuse University, Syracuse, N. Y. 1894.

Prof. EDWARD DELAVAN PERRY (Columbia Univ.), 542 West 114th St.,
New York, N. Y. 1879.

Rev. Dr. JOHN P. PETERS, 225 West 99th St., New York, N. Y. 1882.

Prof. DAVID PHILIPSON, Hebrew Union College, Cincinnati, 0. 1889.

WILLIAM POPPER, 260 West 93d St., New York, N. Y. 1897.

Prof. IBA M. PRICE, Univ. of Chicago, Chicago, 111. 1887.

Prof. JOHN DYNELEY PRINCE (Columbia Univ.), Sterlington, Rockland
Co., N. Y. 1888.

GEORGE PAYN QUACKENBOS, 331 West 28th St., New York, N. Y. 1904.

Pres. F. P. RAMSAY (King College), Bristol, Tenn.

HORACE M. RAMSEY, San Mateo, Cal. 1902.

Dr. HERMANN RANKE, Royal Museum, Berlin, Germany. 1905.

Dr. GEORGE ANDREW REISNER, The Pyramids, Cairo, Egypt. 1891.

Vol. xxvii.] List of Members. 481

ERNEST C. RICHARDSON, Library of Princeton Univ., Princeton, N. J.
1900.

J. NELSON ROBERTSON, 294 Avenue Road, Toronto, Ont. 1902.

BENJ. W. ROBINSON, 700 Park Ave., New York, N. Y. 1906.

EDWARD ROBINSON, Metropolitan Museum of Art, New York, N. Y. 1894.

Prof. GEORGE LIVINGSTON ROBINSON (McCormick Theol. Sem.), 10 Chal-
mers Place, Chicago, 111. 1892.

Hon. WILLIAM WOODVILLE ROCKHILL, Peking, China. 1880.

Prof. ROBERT W. ROGERS, Drew Theological Seminary, Madison, N. J.
1888.

Prof. JAMES HARDY ROPES {Harvard University), 13 Follen St., Cam-
bridge, Mass. 1893.

WILLIAM ROSENAU, Johns Hopkins Univ., Baltimore, Md. 1897.

Miss ADELAIDE RUDOLPH, 18 Wilbur St., Cleveland, O. 1894.

Mrs. JANET E. RUUTZ-REES, 219 West 80th St., New York, N. Y. 1897.

Miss CATHARINE B. RTJNKLE, 15 Everett St., Cambridge, Mass. 1900.

ARTHUR W. RYDER (University of California), 2243 Piedmont Way,
Berkeley, Cal. 1902.

Mrs. EDW. E. SALISBURY, 237 Church St., New Haven, Conn. 1906.

Rev. Dr. FRANK K. SANDERS, 22 Rockview St., Jamaica Plain, Mass.
1897.

President S. SCHECHTER (JeAvish Theological Seminary), 501 West 113th
St., New York, N. Y. 1904.

JOHANN F. SCHELTEMA, care of Messrs. Kerkhoven & Co., 115 Heeren-
gracht, Amsterdam, Holland. 1906.

H. ERNEST SCHMID, White Plains, N. Y. 1866.

Prof. NATHANIEL SCHMIDT, Cornell University, Ithaca, N. Y.

MONTGOMERY SCHUYLER, JR., Dept. of State, Washington, D. C. 1899.

GILBERT C. SCOGGIN, Caruthersville, Mo. 1906.

CHARLES P. G. SCOTT, 150 Woodworth Ave., Yonkers, N. Y. 1895.

Rev. WILLIAM G. SEIPLE, 78 Higashi Sambancho, Sendai, Japan. 1902.

J. HERBERT SENTER, 10 Avon St., Portland, Me. 1870.

CHARLES C. SHERMAN, 65 Irving Place, New York, N. Y. 1904.

fThe Very Rev. JOHN R. SLATTERY, 261 Central Park West, New York,
N. Y. 1903.

Prof. HENRY PRESERVED SMITH, 419 W. 118th "St., New York, N. Y. 1877.

JOHN M. P. SMITH, Univ. of Chicago, 111. 1906.

WILLIAM WALLACE SPENCE, JR., Bolton, Baltimore, Md. 1900.

EDWARD H. SPIEKER, Johns Hopkins Univ., Baltimore, Md. 1884.

Prof. HANS H. SPOER, Ph.D., Theological Seminary, Meadville, Pa. 1899.

Rev. DAVID BRAINARD SPOONER, The Archaeological Survey, Peshawar,
India. 1902.

Prof. CHARLES C. STEARNS, 126 Garden St., Hartford, Conn. 1899.

Rev. JAMES D. STEELE, 15 Grove Terrace, Passaic, N. J. 1892.

Rev. ANSON PHELPS STOKES, JR., Yale University, New Haven, Conn.
1900.

Prof. EDWARD HENRY STROBEL, care Foreign Office, Bangkok, Siam. 1903.

MAYER SULZBERGER, 1303 Girard Ave., Philadelphia, Pa. 1888.
VOL. xxvii. 32

482 American Oriental Society's Proceedings, April, 1906. [1906.

HENRY OSBORN TAYLOB, Century Association, 7 West 43d St., New York,

N. Y. 1899.

EBEN FRANCIS THOMPSON, 311 Main St., Worcester, Mass. 1906.
Rev. Dr. J. J. TIERNEY, Mount St. Mary's College, Emmitsburg, Md.

1901.
Prof. HENBY A. TODD (Columbia University), 824 \Vest End Ave., New

York, N. Y. 1885.

OLAF A. TOFFTEEN, 1113 Washington Building, Chicago, 111.
Prof. CHARLES C. TORREY (Yale University), 67 Mansfield St., New

Haven, Conn. 1891.
Prof. CRAWFORD H. TOY (Harvard University), 7 Lowell St., Cambridge,

Mass. 1871.
Rev. JOSEPH VINCENT TRACY^ 20 Holton St., Allston, Boston, Mass.

1892.
ADDISON VAN NAME (Yale Univ.), 121 High St., New Haven, Conn.

1863.

THOMAS E. WAGGAMAN, 917 F St., N. W., Washington, D. C. 1897.
Miss SUSAN HAYES WAKD, The Stone House, Abington Ave., Newark,

N. J. 1874.

Rev. Dr. WILLIAM HAYES WARD, 130 Fulton St., New York, N. Y. 1869.
Miss CORNELIA WTARREN, Cedar Hill, Waltham, Mass. 1894.
Prof. WILLIAM F. WAJIREN (Boston Univ.), 131 Davis Ave., Brookline,

Mass. 1877.

Rev. W. SCOTT WATSON, West New York, New Jersey. 1893.
Prof. J. E. WERREN, 17 Leonard Ave., Cambridge, Mass. 1894.
Prof. JENS IVERSON WESTENGARD (Harvard Univ.), Asst. Gen. Adviser

to H.S.M. Govt., Bangkok, Siam. 1903.
Pres. BENJAMIN IDE WHEELER, University of California, Berkeley, Cal.

1885.

Prof. JOHN WILLIAMS WHITE (Harvard Univ.), 18 Concord Ave., Cam-
bridge, Mass. 1877.

Miss MARIA WHITNEY, 2 Divinity Ave., Cambridge, Mass. 1897.
Mrs. WILLIAM DWIGHT WHITNEY, 227 Church St., New Haven, Conn.

1897.

Rev. E. T. WILLIAMS, U. S. Legation, Peking, China. 1901.
Prof. FREDERICK WELLS WILLIAMS (Yale Univ.), 135 Whitney Ave.,

New Haven, Conn. 1895.

TALCOTT WILLIAMS ("The Press"), 916 Pine St., Philadelphia, Pa. 1884.
Rev. Dr. WILLIAM COPLEY WINSLOW, 525 Beacon St., Boston, Mass.

1885.

Rev. STEPHEN S. WISE, 46 East 68th St., New York, N. Y. 1894.
HENRY B. WITTON, Inspector of Canals, 16 Murray St., Hamilton,

Ontario. 1885.

Louis B. WOLFENSON, 513 Laurens St., Baltimore, Md. 1904.
WILLIAM W. WOOD, 2802 Parkwood Ave., Baltimore, Md. 1900.
JAMES H. WOODS (Harvard Univ.), 2 Chestnut St., Boston, Mass. 1900.
Prof. JOHN HENRY WRIGHT (Harvard Univ.), 38 Quincy St., Cambridge,

Mass. 1898.
Prof. THEODOBE F. WRIGHT, 42 Quincy St., Cambridge, Mass. 1893.

Vol. xxvii.J List of Members. 483

Rev. JAMES OWENS WRIGHTSON, 1031 Monument St., Baltimore, Md.

1903.
Rev. ABRAHAM YOHANNAN, Columbia University, New York, N. Y. 1894.

1894.

[Total, 273.]

III. MEMBERS OF THE SECTION FOR THE HISTORICAL
STUDY OF RELIGIONS.

Prof. FELIX ADLER, Ph.D., 123 East 60th St., New York, N. Y. 1900.

Rev. Dr. SAMUEL H. BISHOP, 500 West 122d St., New York, N. Y. 1898.

Rev. JOHN L. CHANDLER, Auburndale, Mass. 1899.

SAMUEL DICKSON, 901 Clinton St., Philadelphia, Pa. 1899.

Prof. FRANKLIN H. GIDDINGS (Columbia Univ.), 150 West 79th St., New
York, N. Y. 1900.

Prof. ARTHUR L. GILLETT, Hartford Theological Seminary, Hartford,
Conn. 1898.

CHARLES B. GULICK (Harvard University), 18 Walker St., Cambridge,
Mass. 1899.

Prof. LINDLEY M. KEASBEY, University of Texas, Austin, Texas. 1903.

Prof. GEORGE T. LADD (Yale Univ.), 204 Prospect St., New Haven, Conn.
1898.

Prof. HINCKLEY G. MITCHELL, Ph.D., D.D. (Boston University), 72 Mt.
Vernon St., Boston, Mass. 1900.

WILLIAM W. NEWELL, 54 Garden St., Cambridge, Mass. 1898.

FRED NORRIS ROBINSON, Ph.D. (Harvard Univ.), Longfellow Park, Cam-
bridge, Mass. 1900.

Rev. Dr. MINOT J. SAVAGE, 34th St. and Park Ave., New York, N. Y.
1898.

Rev. W. A. SHEDD, Am. Mission, Urumia, Persia (via Berlin and Tabriz).
1906.

Prof. EDWIN R. SELIGMAN (Columbia Univ.), 324 West 86th St., New
York, N. Y. 1898.

Prof. LANGDON C. STEWARDSON, Lehigh University, South Bethlehem, Pa.
1901.

Prof. WILLIAM G. SUMNER (Yale Univ.), 240 Edwards St., New Haven,
Conn. 1898.

Prof. CHARLES MELLEN TYLER, Cornell Univ., Ithaca, N. Y. 1904.

Prof. R. M. WENLEY, University of Michigan, Ann Arbor, Mich. 1898.

Rev. G. E. WHITE, Anatolia College, Marsovan, Turkey [papers to Ger-
man Consulate (White), Samsoun, Turkey.] 1906.

Prof. IRVING F. WOOD, Smith College, Northampton, Mass. 1905.

[Total, 21.]
Number of Members of all Classes, 319.

SOCIETIES, LIBRARIES, TO WHICH THE PUBLICATIONS OF THE AMERICAN
ORIENTAL SOCIETY ARE SENT BY WAY OF GIFT OR EXCHANGE.

I. AMERICA.

BOSTON, MASS.: American Academy of Arts and Sciences.
CHICAGO, ILL.: Field Museum of Natural History.

484 American Oriental Society's Proceedings: April, 1906. [1906.

NEW YORK: American Geographical Society.
PHILADELPHIA, PA. : American Philosophical Society.

Free Museum of Science and Art, Univ. of Penna.
WASHINGTON, D. C. : Smithsonian Institution.

Bureau of American Ethnology.
WORCESTER, MASS.: American Antiquarian Society.

II. EUROPE.

AUSTRIA, VIENNA: Kaiserliche Akademie der Wissenschaften.

Anthropologische Gesellschaft.

PRAGUE : Kb'niglich Bb'hmische Gesellschaft der Wissenschaften.
DENMARK, ICELAND, REYKJAVIK: University Library.

FRANCE, PARIS : Societe" Asiatique. (Rue de Seine, Palais de 1'Institut.)
Bibliotheque Nationale.
Musee Guimet. (Avenue du Trocade'ro.)
Academic des Inscriptions et Belles-Lettres.
Ecole des Langues Orientales Vivantes. { Rue de Lille, 2. )
GERMANY, BERLIN : Kb'niglich Preussische Akademie der Wissenschaften.
Kbnigliche Bibliothek.
Seminar fur Orientalische Sprachen. (Am Zeug-

hause 1.)
GOTTINGEN: Kbnigliche Gesellschaft der Wissenschaften.

HALLE: Bibliothek der Deutschen Morgenlandischen Gesell-
schaft. ( Friedrichstr. 50.)

LEIPZIG : Kbniglich Sachsische Gesellschaft der Wissenschaften.

Leipziger Semitistische Studien. (J. C. Hinrichs.)

MUNICH : Kbniglich Bairische Akademie der Wissenschaften.

Kbnigliche Hof- und Staatsbibliothek.
TUBINGEN: Library of the University.
GREAT BRITAIN, LONDON: Royal Asiatic Society of Great Britain and

Ireland. (22 Albemarle St., W.)
Library of the India Office. (Whitehall, SW.)
Society of Biblical Archaeology. (37 Great

Russell St., Bloomsbury, W.C.)
Philological Society. (Care of Dr. F. J.
Furnival, 3 St. George's Square, Prim-
rose Hill, NW.)
ITALY, FLORENCE: Societa Asiatica Italiana.

ROME: Reale Accademia dei Lincei.

NETHERLANDS, AMSTERDAM : Koninkli jke. Akademie van Wetenschappen.
THE HAGUE: Koninklijk Instituut voor Taal-, Land-, en

Volkenkunde van Nederlandsch Indie.
LEYDEN: Curatorium of the University.
RUSSIA, HELSINGFORS: Societe" Finno-Ougrienne.

ST. PETERSBURG: Imperatorskaja Akademija Nauk.

Archeologiji Institut.
SWEDEN, UPSAXA: Humanistiska Vetenskaps-Samfundet.

Le Monde Oriental (cf. Professor K. F. Johansson,
Upsala ) .

Vol. xxvii. List of Members. 485

III. ASIA.

CALCUTTA, GOV'T OF INDIA: Home Department.

CEYLON, COLOMBO: Ceylon Branch of the Royal Asiatic Society.

CHINA, SHANGHAI: China Branch of the Royal Asiatic Society.

TONKIN: 1'Ecole Franchise d' extreme Orient (Rue de Coton),

Hanoi.

INDIA, BOMBAY: Bombay Branch of the Royal Asiatic Society.
The Anthropological Society. (Town Hall.)
CALCUTTA: The Asiatic Society of Bengal. (57 Park St.)

The Buddhist Text Society. (86 Jaun Bazar St.)
LAHORE: Library of the Oriental College.

SIMLA: Office of the Director General of Archaeology. (Ben-
more, Simla, Punjab.)

JAPAN, TOKYO: The Asiatic Society of Japan.

JAVA, BATAVIA : Bataviaasch Genootschap van Kunsten en Wetenschappea.
KOREA: Branch of Royal Asiatic Society, Seoul, Korea.
NEW ZEALAND: The Polynesian Society, New Plymouth.
PHILIPPINE ISLANDS: The Ethnological Survey, Manila.
SYRIA: The American School (care U. S. Consul, Jerusalem).
Revue Biblique, care of M. J. Lagrange, Jerusalem.
Al-Machriq, The Catholic Press, Beirut, Syria.

IV. AFRICA.
EGYPT, CAIRO: The Khedivial Library.

V. EDITORS OF THE FOLLOWING PERIODICALS.

The Indian Antiquary (Education Society's Press, Bombay, India).
Wiener Zeitschrift fur die Kunde des Morgenlandes (care of Alfred

Holder, Rothenthurm-str. 15, Vienna, Austria).
Zeitschrift fiir vergleichende Sprachforschung (care of Prof. E. Kuhn, 3

Hess Str., Munich, Bavaria) .
Revue de PHistorie des Religions (care of M. Jean ReVille, chez M. E.

Leroux, 28 rue Bonaparte, Paris, France).

Zeitschrift fiir die alttestamentliche Wissenschaft (care of Prof. Bern-
hard Stade, Giessen, Germany ) .
Beitrage zur Assyriologie und semitischen Sprachwissenschaft. (J. C.

Hinrichs'sche Buchhandlung, Leipzig, Germany.)
Oriental Bibliography (care of Prof. Lucian Scherman, 18 Ungerer Str.,

Munich, Bavaria).
The American Antiquarian and Oriental Journal, 438 East 57th St.,

Chicago, 111.
American Journal of Archaeology, 65 Sparks St., Cambridge, Mass.

RECIPIENTS: 319 (Members) -f 69 (Gifts and Exchanges) — 388.

486 American Oriental Society's Proceedings, April, 1906. [1906.

REQUEST.

The Editors request the Librarians of any Institution or Libraries, not
mentioned above, to which this Journal may regularly come, to notify
them of the fact. It is the intention of the Editors to print a list, as
complete as may be, of regular subscribers for the Journal or of recipients
thereof. The following is the beginning of such a list.

Andover Theological Seminary.

Boston Public Library.

Brown University Library.

Chicago University Library.

Columbia University Library.

Cornell University Library.

Harvard Sanskrit Class-Room Library.

Harvard Semitic Class-Room Library.

Harvard University Library.

Nebraska University Library.

New York Public Library.

Yale University Library.

Vol. xxvii.] Constitution and By-Laios. 487

CONSTITUTION AND BY-LAWS

OF THE

AMERICAN ORIENTAL SOCIETY.

With Amendments of April, 1897.

CONSTITUTION.

ARTICLE I. This Society shall be called the AMERICAN ORIENTAL SOCIETY.
ARTICLE II. The objects contemplated by this Society shall be: —

1. The cultivation of learning in the Asiatic, African, and Polynesian
languages, as well as thte encouragement of researches of any sort by
which the knowledge of the East may be promoted.

2. The cultivation of a taste for oriental studies in this country.

3. The publication of memoirs, translations, vocabularies, and other
communications, presented to the Society, which may be valuable with
reference to the before-mentioned objects.

4. The collection of a library and cabinet.

ABTICLE III. The members of this Society shall be distinguished as
corporate and honorary.

ARTICLE IV. All candidates for membership must be proposed by the
Directors, at some stated meeting of the Society, and no person shall be
elected a member of either class without receiving the votes of as many as
three-fourths of all the members present at the meeting.

ARTICLE V. The government of the Society shall consist of a President,
three Vice Presidents, a Corresponding Secretary, a Recording Secretary,
a Secretary of the Section for the Historical Study of Religions, a
Treasurer, a Librarian, and seven Directors, who shall be annually elected
by ballot, at the annual meeting.

ARTICLE VI. The President and Vice Presidents shall perform the
customary duties of such officers, and shall be ex-officio members of the
Board of Directors.

ARTICLE VII. The Secretaries, Treasurer, and Librarian shall be
ex-officio members of the Board of Directors, and shall perform their
respective duties under the superintendence of said Board.

ARTICLE VIII. It shall be the duty of the Board of Directors to regu-
late the financial concerns of the Society, to superintend its publications,
to carry into effect the resolutions and orders of the Society, and to
exercise a general supervision over its affairs. Five Directors at any
regular meeting shall be a quorum for doing business.

ARTICLE IX. An Annual meeting of the Society shall be held during
Easter week, the days and place of the meeting to be determined by the
Directors, said meeting to be held in Massachusetts at least once in three

488 American Oriental Society's Proceedings, April, 1906. [1906.

years. One or more other meetings, at the discretion of the Directors,
may also be held each year at such place and time as the Directors shall
determine.

ARTICLE X. There shall be a special Section of the Society, devoted to
the historical study of religions, to which section others than members of
the American Oriental Society may be elected in the same manner as is
prescribed in Article IV.

ABTICLE XT. This Constitution may be amended, on a recommendation
of the Directors, by a vote of three-fourths of the members present at an
annual meeting.

BY-LAWS.

I. The Corresponding Secretary shall conduct the correspondence of
the Society, and it shall be his duty to keep, in a book provided for the
purpose, a copy of his letters; and he shall notify the meetings in such
manner as the President or the Board of Directors shall direct.

II. The Recording Secretary shall keep a record of the proceedings of
the Society in a book provided for the purpose.

III. a. The Treasurer shall have charge of the funds of the Society;
and his investments, deposits, and payments shall be made under the
superintendence 'of the Board of Directors. At each annual meeting he
shall report the state of the finances, with a brief summary of the receipts
and payments of the previous year.

III. 6. After December 31, 1896, the fiscal year of the Society shall
correspond with the calendar year.

III. c. At each annual business meeting in Easter week, the President
shall appoint an auditing committee of two men — preferably men residing
in or near the town where the Treasurer lives — to examine the Treasurer's
accounts and vouchers, and to inspect the evidences of the Society's prop-
erty, and to see that the funds called for by his balances are in his hands.
The Committee shall perform this duty as soon as possible after the New
Year's day succeeding their appointment, and shall report their findings
to the Society at the next annual business meeting thereafter. If these
findings are satisfactory, the Treasurer shall receive his acquittance by a
certificate to that effect, which shall be recorded in the Treasurer's book,
and published in the Proceedings.

IV. The Librarian shall keep a catalogue of all books belonging to the
Society, with the names of the donors, if they are presented, and shall at
each annual meeting make a report of the accessions to the library during
the previous year, and shall be farther guided in the discharge of his
duties by such rules as the Directors shall prescribe.

V. All papers read before the Society, and all manuscripts deposited
by authors for publication, or for other purposes, shall be at the disposal
of the Board of Directors, unless notice to the contrary is given to the
Editors at the time of presentation.

VI. Each corporate member shall pay into the treasury of the Society
an annual assessment of five dollars; but a donation at any one time of
seventy-five dollars shall exempt from obligation to make this payment.

Vol. xxvii.] Constitution and By-Laws. 489

VII. Corporate and Honorary members shall be entitled to a copy of
all the publications of the Society issued during their membership, and
shall also have the privilege of taking a copy of those previously pub-
lished, so far as the Society can supply them, at half the ordinary selling
price.

VIII. Candidates for membership who have been elected by the
Society shall qualify as members by payment of the first annual assess-
ment within one month from the time when notice of such election is
mailed to them. A failure so to qualify shall be construed as a refusal
to become a member. If any corporate member shall for two years fail
to pay his assessments, his name may, at the discretion of the Directors,
be dropped from the list of members of the Society.

IX. Members of the Section for the Historical Study of Religions
shall pay into the treasury of the Society an annual assessment of two
dollars; and they shall be entitled to a copy of all printed papers which
fall within the scope of the Section.

X. Six members shall form a quorum for doing business, and three
to adjourn.

SUPPLEMENTARY BY-LAWS.

I. FOB THE LlBBAEY.

1. The Library shall be accessible for consultation to all members of
the Society, at such times as the Library of Yale College, with which it is
deposited, shall be open for a similar purpose; further, to such persons
as shall receive the permission of the Librarian, or of the Librarian or
Assistant Librarian of Yale College.

2. Any member shall be allowed to draw books from the Library upon
the following conditions: he shall give his receipt for them to the
Librarian, pledging himself to make good any detriment the Library may
suffer from their loss or injury, the amount of said detriment to be
determined by the Librarian, with the assistance of the President, or of
a Vice President; and he shall return them within a time not exceeding
three months from that of their reception, unless by special agreement
with the Librarian this term shall be extended.

3. Persons not members may also, on special grounds, and at the
discretion of the Librarian, be allowed to take and use the Society's books,
upon depositing with the Librarian a sufficient security that they shall
be duly returned in good condition, or their loss or damage fully com-
pensated.

PUBLICATIONS OF THE AMERICAN ORIENTAL

SOCIETY.

PRICE OF THE JOURNAL.

Vol. I. (1843-1849) No. 1 (Nos. 2-4 out of print) $ .50

Vol. II. (1851) - 2.50

Vol. III. (1852-1853) 2.50

Vol. IV. (1853-1854) 2.50

Vol. V. (1855-1856) 2.50

Vol. VI. (1860) 5.00

Vol. VII. (1862) 5.00

Vol. VIII. (1868) 5.00

Vol. IX. (1871) 5.00

Vol. X. (1872-1880) 6.00

Vol. XL (1882-1885) 5.00

Vol. XII. (1881) ..'. 4.00

Vol. XIII. (1889) 6.00

Vol. XIV. (1890). 5.00

Vol. XV. (1893) 5.00

Vol. XVI. (1894-1896).. 5.00

Vol. XVII. (1896) bound in full buckram 2.50

Vol. XVIII. First and Second Half (1897) buckram, each 2.50 5.00

Vol. XIX. First Half (1898) full cloth 1.50

Vol. XIX. Second Half (1898) bound in full buckram 2. 50

Vol. XX. First and Second Half (1899) buckram, each 2.50 5.00

Vol. XXI. First Half (Index) 2.50

Vol. XXI. Second Half (1900) bound in full buckram 2.50

Vol. XXII. First and Second Half (1901) buckram, each 2.50 5.00

Vol. XXIII. First and Second Half (1902) buckram, each 2.50 5.00

Vol. XXIV. First and Second Half (1903) buckram, each 2.50 5.00

Vol. XXV. First and Second Half (1904) buckram, each 2.50 5.00

Vol. XXVI. First and Second Half (1905) buckram, each 2.50 5.00

Vol. XXVII. First and Second Half (1906) buckram, each 2.50 5.00

Total $118.00

Whitney's Taittiriya-Praticakhya (vol. ix.) $5.00

Avery's Sanskrit Verb-Inflection (from vol. x.) 75

Whitney's Index Verborum to the Atharva-Veda (vol. xii.) 4.00

The same (vol. xii.) on large paper 5.00

Hopkins's Position of the Ruling Caste (from vol. xiii.) 3.00

Oertel's Jaiminiya-Upanisad-Brahmana (from vol. xvi.) 1.75

Arnold's Historical Vedic Grammar (from vol. xviii.) 1.75

Bloomfield's Kauc.ika-Su.tra of the Atharva-Veda (vol. xiv.) 5.00

The Whitney Memorial Volume (vol. xix. , first half) 1.50

For any of the above, address the Librarian of the Society, Professor
Hanns Oertel, New Haven, Connecticut. Members can have the series
at half price. To public libraries or those of educational institutions,
Vol. I. No. 1 and Vols. II. to V. will be given free, and the rest sold
at a discount of twenty per cent.

492 Notices.

TO CONTRIBUTORS.

Fifty copies of each article published in this Journal will be
forwarded to the author. A larger number will be furnished at
cost.

Arabic, Persian, Syriac, (Jacobite and Nestorian), Armenian,
Coptic, Ethiopic, Sanskrit, Tamil, Chinese, and Japanese fonts
of types are provided for the printing of the Journal, and others
will be procured from time to time, as they are needed.

GENERAL NOTICES.

1. Members are requested to give immediate notice of changes
of address to the Treasurer, Prof. Frederick Wells Williams,
135 Whitney avenue, New Haven, Conn.

2. It is urgently requested that gifts and exchanges intended
for the Library of the Society be addressed as follows: The
Library of the American Oriental Society, Yale University,
New Haven, Connecticut, IT. S. America.

3. For information regarding the sale of the Society's pub-
lications, see the next foregoing page.

4. Communications for the Journal should be sent to Prof.
E. Washburn Hopkins or Prof. Charles C. Torrey, New Haven.

CONCERNING MEMBERSHIP.

It is not necessary for any one to be a professed Orientalist in
order to become a member of the Society. All persons — men or
women — who are in sympathy with the objects of the Society
and willing to further its work are invited to give it their help.
This help may be rendered by the payment of the annual assess-
ments, by gifts to its library, or by scientific contributions to its
Journal, or in all of these ways. Persons desiring to become
members are requested to apply to the Treasurer, whose address
is given above. Members receive the Journal free. The
annual assessment is $5. The fee for Life-Membership is $75.

Persons interested in the Historical Study of Religion may
become members of the Section of the Society organized for this
purpose. The annual assessment is $2 ; members receive copies
of all publications of the Society which fall within the scope of
the Section.

PJ

2

A5

v.27
pt.2

American Oriental Society
Journal

PLEASE DO NOT REMOVE
CARDS OR SLIPS FROM THIS POCKET

UNIVERSITY OF TORONTO LIBRARY